JP2016031640A - Virtual machine management apparatus, virtual machine management system, virtual machine management method, and virtual machine management program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a management device, a management system, a management method and a management program, capable of surely moving a virtual machine.SOLUTION: In a management device 12, a relay point calculation function unit 14 calculates a relay point (100), and a virtualization resource control function unit 17 acquires resource information of a virtual machine of a movement object from a data center DC1 of a movement source (202). The management device requires securement of a resource with the acquired resource information of the virtual machine of the movement object, at the virtualization resource control function part (104). In a data center DC2 of the relay point and a data center DC3 of a movement destination, according to the requirement from the management device, the resource of the virtual machine of the movement target is secured in a usable manner. Consequently, before movement of the virtual machine, a resource that the virtual machine uses can be secured.SELECTED DRAWING: Figure 14

Description

  The disclosed technology relates to a virtual machine management apparatus, a virtual machine management system, a virtual machine management method, and a virtual machine management program.

  Realization of virtualization technology to build multiple virtual computers (hereinafter referred to as virtual machines) on a single physical machine for the purpose of improving the utilization efficiency of physical computers (hereinafter referred to as physical machines) Has been. In recent years, it has become common to perform business processing using physical machines including virtual machines based on virtualization technology.

  A virtual machine built on a physical machine can be moved to another physical machine. For example, a technique called live migration that moves a virtual machine to another physical machine without stopping a business process that is running on the virtual machine constructed on a specific physical machine is known. When the live migration technology is applied, a virtual machine running on a certain physical machine that is executing a business process is transferred to another physical machine.

  When moving a virtual machine on a source physical machine to another physical machine that is a destination, it is required to move the virtual machine reliably without any problems. For example, in the live migration technology, it is known that physical machine resources are consumed when executing a process of moving a virtual machine. If the movement of the virtual machine is started when the resources of the physical machine are saturated, the service provided by the physical machine may be reduced depending on the load due to live migration. In order to suppress a decrease in service provided by a physical machine, a technique is known in which resources used for moving a virtual machine are secured in a source physical machine when the virtual machine is moved.

  To ensure the movement of the virtual machine, set the destination physical machine to which the virtual machine can be moved, monitor the operating status of the virtual machine, and move the virtual machine. A technique for selecting a machine and a moving route is known.

JP 2011-090537 A JP 2011-232916 A

  By the way, when moving a virtual machine, there is a case where a physical machine other than the source physical machine and the destination physical machine relays between the source physical machine and the destination physical machine. For example, when a virtual machine is moved over a long distance from the source physical machine to the destination physical machine, the physical path to be relayed is selected based on the response time of the physical machine, and the transfer route is set. The virtual machine is moved according to the movement route. When a virtual machine is moved from a source to a destination via a relay point without stopping the business process that is running on the virtual machine, the virtual machine may be operable on each destination physical machine desirable.

  However, there may be a problem in the movement of the virtual machine due to a shortage of resources for operating the virtual machine in the destination physical machine and a problem in the operation of the virtual machine. In other words, the virtual machine may not move from the source physical machine to the destination physical machine via the relay destination physical machine.

  In one aspect, an object is to reliably move a virtual machine from a source physical machine to a destination physical machine.

  In the disclosed technology, a virtual machine to be moved that operates on the first physical machine that is the movement source is relayed by the second physical machine that is the relay destination, and is moved to the third physical machine that is the movement destination. . The acquisition unit acquires resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the first physical machine. Further, the acquisition unit acquires position information of each of the first physical machine, the second physical machine, and the third physical machine, and movement order information of the physical machine to which the virtual machine to be moved is moved. The request unit secures the resource allocation amount indicated by the resource allocation information of the virtual machine to be migrated to each of the second physical machine and the third physical machine based on the position information acquired by the acquisition unit. The request information shown is transmitted. The instruction unit receives the response of the request information transmitted by the request unit from each of the second physical machine and the third physical machine, and based on the position information acquired by the acquisition unit, the movement order information, and the received response, An instruction is given to move the virtual machine to be moved according to the movement order information.

  In one embodiment, the virtual machine can be reliably moved from the source physical machine to the destination physical machine.

1 is a block diagram illustrating a virtual machine management system according to a first embodiment. FIG. It is an image figure which shows an example of a virtualization infrastructure table. It is an image figure which shows an example of a virtual machine table. It is an image figure which shows an example of a virtual infrastructure utilization rate information table. It is an image figure which shows an example of a virtual resource empty block management table. It is an image figure which shows an example of the VM time limit timer table which concerns on the virtual resource timer management table. It is an image figure which shows an example of the time limit value table which concerns on the virtualization resource timer management table. It is an image figure which shows an example of a relay point list. FIG. 6 is an explanatory diagram for schematically explaining a usage rate of a CPU. It is explanatory drawing for demonstrating and explaining the usage rate of a memory. It is a block diagram of the computer system which can implement | achieve a virtual machine management system. It is a flowchart which shows an example of the flow of a process in a management apparatus. It is a flowchart which shows an example of the flow of a process of the virtualization infrastructure program performed in a data center. FIG. 11 is a sequence diagram illustrating an example of information exchange regarding resource reference and resource reservation of a virtual machine. It is a flowchart which shows an example of the flow of the information acquisition process performed with a management apparatus. It is a flowchart which shows an example of the flow of the resource reference process performed in a data center. It is a flowchart which shows an example of the flow of the process which requests | requires resource reservation performed with a management apparatus. It is a flowchart which shows an example of the flow of the resource securing process performed in a data center. It is a sequence diagram showing an example of information exchange in live migration. It is a flowchart which shows an example of the flow of the movement process performed with a management apparatus. It is a flowchart which shows an example of the flow of the movement process performed in a data center. It is a sequence diagram which shows an example of information transfer about the resource release of a virtual machine. It is a flowchart which shows an example of the flow of a request process of the resource release performed with a management apparatus. It is a flowchart which shows an example of the flow of the resource release process performed in a data center. It is a sequence diagram showing an example of information exchange in the virtual machine management system according to the embodiment. It is a flowchart which shows an example of the flow of the information acquisition process which concerns on 2nd Embodiment. It is a flowchart which shows an example of the flow of the resource reference process which concerns on 2nd Embodiment. It is a flowchart which shows an example of the flow of the resource securing request process which concerns on 2nd Embodiment. It is a flowchart which shows an example of the flow of the resource securing process which concerns on 2nd Embodiment. It is a sequence diagram which shows an example of information transfer in the live migration which concerns on 2nd Embodiment. It is a flowchart which shows an example of the flow of the movement process performed with the management apparatus which concerns on 2nd Embodiment. It is a sequence diagram which shows an example of information transfer in the virtual machine management system which concerns on 3rd Embodiment. It is an image figure which shows an example of the virtual infrastructure utilization rate information table which concerns on 3rd Embodiment. It is an image figure which shows an example of the virtual resource empty block management table which concerns on 3rd Embodiment.

  Hereinafter, an example of an embodiment of the disclosed technology will be described in detail with reference to the drawings.

<First Embodiment>
FIG. 1 shows an example of the configuration of a virtual machine management system 10 according to the present embodiment. In the virtual machine management system 10, a management apparatus 12 and a plurality of data centers including virtual machines are connected to a network 20 such as a management network. FIG. 1 shows an example in which the virtual machine management system 10 includes three data centers 22 (in FIG. 1, DC1, DC2, and DC3 are indicated internally). In FIG. 1, an example in which the virtual machine management system 10 includes three data centers 22 is shown, but the number is not limited to three, and may be four or more.

  The management device 12 is for managing resources (hereinafter referred to as resources) used by virtual machines in the data center 22 in relation to execution of live migration, and can be realized by, for example, a computer (details will be described later). ). The management device 12 includes an information collection management function unit 13, a relay point calculation function unit 14, a migration management function unit 15, a virtualized resource control function unit 17, and a storage unit 16. The management device 12 can include a virtual resource management function unit 18. In addition, the storage unit 16 holds physical machine and virtual machine states as various types of management information.

  The information collection management function unit 13 is a function unit that collects and manages various information such as IP addresses (Internet Protocol Address) of various nodes connected to the network 20 and delay times between the nodes. The various nodes include other management devices different from the management device 12, physical machines provided in the data center, virtual machines operating on the physical machines, and network devices connected to the physical machines and virtual machines. For example, the information collection management function unit 13 collects information on the virtualization infrastructure on the virtual machine management system 10, information on the virtual machine, and the like, and stores and manages the information in the storage unit 16. Note that the information regarding the virtual machine includes information indicating the virtual machine and resource information indicating the resource used in the virtual machine. For example, the resource information includes the name of the resource and the usage amount (allocation amount) for the resource used in the virtual machine.

  The relay point calculation function unit 14 is a functional unit that calculates a relay point from the source data center to the destination data center when performing live migration in accordance with an instruction from an administrator or the like. The relay point calculation function unit 14 moves the virtual machine to be moved by live migration, and relays from the source to the destination by inputting information indicating the source data center 22 and the destination data center 22. And the route (order) is calculated.

  The relay point calculation function unit 14 may hold information indicating the calculated relay point and route (order) to be relayed from the movement source to the movement destination, or may be stored in the storage unit 16. When information indicating the relay point and the route (order) is stored in the storage unit 16, the relay point calculation function unit 14 sends information to the information collection management function unit 13, and the information collection management function unit 13 stores the information in the storage unit 16. May be.

  The migration management function unit 15 is a function unit that manages information and processing related to live migration, and includes an interface function that exchanges information with other management devices connected to the network 20 and a data center. For example, the migration management function unit 15 instructs the data center 22 to move the virtual machine 28 according to the relay point and route (order) calculated by the relay point calculation function unit 14 from the movement source to the movement destination.

  The virtualization resource control function unit 17 is a function unit that instructs the virtualization resource management function unit 34 that manages virtual machines in the data center 22 to control the securing and releasing of resources used by the virtual machine 28. is there. In other words, the virtualization resource control function unit 17 instructs the data center 22 to secure resources used by the virtual machine 28 so that the virtual machine 28 is reliably moved from the movement source to the movement destination via the relay point. Or give instructions to release resources that have been used.

  Note that the management apparatus 12 may include the virtual resource management function unit 18 and manage virtual machines in the data center 22.

  The data center 22 connected to the network 20 can construct one or a plurality of virtual machines on one or a plurality of physical machines using a virtualization technique. In the data center 22, each of one or a plurality of virtual machines to be constructed is managed. In this embodiment, the data center 22 is described as one physical machine for the sake of simplicity.

  In the following description, when the data center 22 at three sites or the elements included in the data center 22 are individually targeted, symbols (DC1, DC2, DC3 shown in FIG. 1) are assigned and distinguished. In addition, as a data center 22 when moving a virtual machine, a data center 22DC1 of a movement source, a data center 22DC2 of a relay site, and a data center 22DC3 of a movement destination will be described as an example.

  The data center 22 includes a gateway (GW) 24, a virtualization infrastructure 26, and a virtual machine 28. The GW 24 mediates information to a destination, and includes a GW / router function unit 30 connected to the network 20. The GW / router function unit 30 is a function unit having a function of a connected device called a router. The GW / router function unit 30 includes various functions such as measuring a delay time with another data center 22 according to an instruction from the management device 12.

  The virtualization infrastructure 26 is a functional unit for managing virtual machines and realizing live migration. The virtualization infrastructure 26 includes a virtualization infrastructure main function unit 32, a virtualization resource management function unit 34, and a virtualization resource control function unit 36.

  The virtualization infrastructure main function unit 32 includes main functions for constructing and operating a virtual machine on a physical machine using a virtualization technique. For example, the virtualization infrastructure main function unit 32 includes an event notification function for notifying the management apparatus 12 of various information during execution of live migration in response to an event such as live migration.

  The virtual resource management function unit 34 includes a function for managing each process of securing and releasing resources used by the virtual machine 28 managed in the data center 22 in accordance with an instruction from the management device 12.

  The virtual resource control function unit 36 performs functions such as securing and releasing resources used by a virtual machine in response to events such as securing and releasing resources used by the virtual machine 28 and executing live migration. including.

  The virtual machine 28 is a virtual computer by operating a virtual machine function unit 38 including an OS (Operating System) operating on the virtualization platform 26 and an application program. In addition, the virtual machine 28 is live migrated onto the virtualization infrastructure 26 of another data center 22 according to an instruction from the management apparatus 12.

  In the present embodiment, the management information stored in the storage unit 16 is used in the virtual machine management system 10 in order to execute a process for reliably performing live migration of the virtual machine 28. As an example of management information, information stored in each of the virtualization infrastructure table 63, the virtual machine table 64, the virtualization infrastructure operating rate information table 65, the virtualization resource empty / busy management table 66, and the virtualization resource timer management table 67 is shown. is there. In the present embodiment, the live migration process is executed with reference to the information in the relay point list 68 (FIG. 8).

  FIG. 2 shows an example of the virtualization infrastructure table 63. In the virtualization infrastructure table 63, information indicating the virtualization infrastructure 26 included in each of all the data centers 22 connected to the network 20 and information indicating the last update date / time are registered as management information. In the example of FIG. 2, the virtualization infrastructure table 63 includes information indicating each of “HV ID”, “HV Name”, “HV IP”, “Site ID”, “Latency”, and “Latest”.

  “HV ID” in the virtualization infrastructure table 63 illustrated in the example of FIG. 2 is identification information unique to the virtualization infrastructure 26. “HV Name” is information indicating the name of the virtualization infrastructure 26. “HV IP” is information indicating the position of an IP address or the like assigned to the virtualization platform 26 as physical machine position information. “Site ID” is identification information of the data center 22 which is a physical machine including the virtualization infrastructure 26. “Latency” is information indicating the time required to access the virtualization platform 26. “Latest” is information indicating the latest date and time when information indicating “HV ID”, “HV Name”, “Site ID”, and “Latency” is acquired and registered or updated.

  FIG. 3 shows an example of the virtual machine table 64. In the virtual machine table 64, information indicating the virtual machines 28 included in each of all the data centers 22 connected to the network 20 and resource information of the virtual machines 28 are registered as management information. In the example of FIG. 3, the virtual machine table 64 includes “VM ID”, “VM Name”, “CPU”, “Memory”, “HDD”, “NIC”, “NW-Name”, and “HV ID”. Contains information indicating each.

  “VM ID” in the virtual machine table 64 illustrated in the example of FIG. 3 is identification information unique to the virtual machine 28. “VM Name” is information indicating the name of the virtual machine 28. “CPU”, “Mem”, and “HDD” are examples of main resource information of the virtual machine 28. “CPU” indicates the usage rate of the CPU, “Mem” indicates the usage amount of the memory, and “HDD” indicates the usage amount of the secondary storage medium such as a hard disk device. “NIC” is information indicating the number of ports connected to the network 20. “NW-Name” is information indicating a network in which the virtual machine 28 participates. “HV ID” is identification information unique to the virtualization infrastructure 26 in charge of the virtual machine 28.

  FIG. 4 shows an example of the virtual infrastructure operating rate information table 65. Information in which information indicating “resource ID” is associated with each piece of information shown in the virtual machine table 64 (FIG. 3) is registered as management information in the virtualization platform operation rate information table 65. The “resource ID” in the virtualization platform operation rate information table 65 illustrated in the example of FIG. 4 is identification information for identifying the resource used by the virtual machine 28 for each virtual machine 28.

  FIG. 5 shows an example of the virtual resource empty / busy management table 66. Information including information indicating the allocation status of resources used by the virtual machine 28 is registered in the virtualized resource empty / busy management table 66 as management information. In the example of FIG. 5, the virtualized resource empty / busy management table 66 includes information indicating “resource ID”, “empty / busy state”, “HV ID”, and “Latest”.

  The “resource ID” of the virtual resource empty / busy management table 66 shown in the example of FIG. 5 is identification information for identifying the resource used by the virtual machine 28, and the virtualization infrastructure operating rate information table through the “resource ID”. 65 can be referred to. The “empty state” is information indicating an allocation state of resources used by the virtual machine 28. As an example of information indicating the allocation status stored as an empty status, “idle” in the idle status, “busy” in the used status, “temporary” in the secured status, and in the migration status The value of “temporary on migration” is stored. “HV ID” is identification information unique to the virtualization platform 26 in charge of the virtual machine 28, and the virtual machine table 64 can be referred to through “HV ID”. “Latest” is information indicating the latest date and time when information indicating “resource ID”, “empty state”, and “HV ID” is acquired and registered or updated.

  6 and 7 show an example of the virtual resource timer management table 67. FIG. The virtual resource timer management table 67 is a time table for determining a time for securing resources used by the virtual machine 28 when not in use. In the present embodiment, a VM timed timer table 67A and a time limit value table 67B are used as the virtualized resource timer management table 67. In the VM time limit timer table 67A, information indicating “resource ID” is associated with information indicating “time limit value”. In the time limit value table 67B, information indicating “resource class” is associated with information indicating “time limit value”. The “resource class” is information indicating a class of resources that are classified in advance according to the type and size of the resources used by the virtual machine 28.

  In the present embodiment, for each resource used by the virtual machine 28, a time limit time for securing the resource is determined. For example, a case where the resource with the resource ID “VRSid007” is classified into the resource class “VRS_class0” will be described. For the resource with the resource ID “VRSid007”, the time limit value “T001” of the resource class “VRS_class0” is specified in the time limit value table 67B shown in FIG. The identified time limit value “T001” is registered in the VM time limit timer table 67A in association with the resource ID “VRSid007”.

  FIG. 8 shows an example of the relay location list 68. The relay point list 68 is a list of information indicating relay points and routes (order) relayed from the movement source to the movement destination calculated by the relay point calculation function unit 14. The example illustrated in FIG. 8 illustrates a case where the virtual machine 28 stays at four data centers 22 on a route along which the virtual machine 28 moves from the movement source to the movement destination. Each of the first HV to the fourth HV stores information indicating the virtualization infrastructure 26 of the data center 22. Further, in the example shown in FIG. 8, three types of relay locations and routes are provided from the source virtualization platform 26 indicated by “H0001” to the destination virtualization platform 26 indicated by “H0003”. Indicates that the route has been calculated. The example shown in FIG. 8 shows the case where the virtual machine 28 is located in four data centers 22, but the number is not limited to four, and may be three or five or more. Good. In addition, although a case where different types of routes are calculated at four locations is shown, the number of virtual machines 28 is not fixed, and three or more data centers 22 are connected to virtual machines for each route. 28 should just be located. Further, the relay point list 68 from which a plurality of types of routes are calculated can be given priority according to the information delay time between the data centers 22 and the like, and can be listed from the one with the highest priority.

  Here, securing of resources used by the virtual machine 28 will be described. In this embodiment, when performing live migration, the resource used by the virtual machine 28 is secured and then live migration is performed.

  FIG. 9 shows an image diagram of the CPU resource 33 in which the CPU usage rate is schematically illustrated as a resource used by the virtual machine 28. The example of FIG. 9 shows that when the virtual performance of the CPU by the virtualization technology is 40 GHz, the CPU resource 33 that is the usage rate is managed in a cell of 2 GHz unit. For example, consider a case where the first virtual machine 28A and the second virtual machine 28B are operating in the migration source data center 22A. When the first virtual machine 28A uses 6 GHz with a CPU usage rate of 30%, the virtualization resource management function unit 34A of the data center 22A of the migration source manages the use of the three cells 33A. When the second virtual machine 28B uses 4 GHz with a CPU usage rate of 20%, the virtual resource management function unit 34A manages the use of the two cells 33B.

  FIG. 10 shows an image diagram of the memory resource 35 in which the memory usage rate is schematically shown as a resource used by the virtual machine 28. The example of FIG. 10 shows that when the virtual performance of the memory by the virtualization technology is 80 GB, the memory resource 35 that is the usage amount is managed in a cell of 4 GB unit. For example, when the memory usage of the first virtual machine 28A is 8 GB, the virtualization resource management function unit 34A of the migration source data center 22A manages the use of the two cells 35A. When the memory usage of the second virtual machine 28B is 12 GB, the virtual resource management function unit 34A manages the use of the three cells 35B.

  When any one of the first virtual machine 28A and the second virtual machine 28B is the migration target virtual machine 28 during the live migration, the resource information of the migration target virtual machine 28 is the virtualization resource control of the management apparatus 12. Acquired by the function unit 17. Here, the CPU resource 33 and the memory resource 35 are acquired. The virtualization resource control function unit 17 of the management apparatus 12 instructs the virtual resource management function units 34B and 34C of the relay destination data center 22B and the migration destination data center 22C to secure resources. Each of the virtualization resource management function units 34B and 34C of the relay destination data center 22B and the migration destination data center 22C responds to an instruction from the virtualization resource control function unit 17 of the management apparatus 12 with a virtualization resource control function. The unit 36B, 36C is instructed to secure resources. Each of the virtual resource control function units 36B and 36C executes resource reservation.

  The virtual machine management system 10 is an example of a virtual machine management system in the disclosed technique, and the management device 12 is an example of a virtual machine management apparatus in the disclosed technique. The information collection management function unit 13 is an example of an acquisition unit in the disclosed technology, and the virtual resource control function unit 17 is an example of a request unit and an instruction unit in the disclosed technology.

  The virtual machine management system 10 can be realized by a computer system 40 shown in FIG. The management device 12 can be realized by, for example, the computer 42 shown in FIG. The data center 22 can be realized by the physical machine 70 shown in FIG.

  When the management device 12 is realized by the computer 42, the computer 42 includes a CPU 44, a memory 46, and a nonvolatile storage unit 54. The CPU 44, the memory 46, and the storage unit 54 are connected to each other via a bus 53. The computer 42 includes a display 48 that is a display device, and a keyboard 49 and a mouse 50 that are input devices. The display 48, the keyboard 49, and the mouse 50 are connected to a bus 53. Further, the computer 42 includes a device (RW device) 51 for reading and writing from / to the recording medium 69, and the RW device 51 is connected to the bus 53. The computer 42 also includes a communication unit (I / O) 52 including an interface for connecting to the network 20. The storage unit 54 can be realized by an HDD (Hard Disk Drive), a flash memory, or the like.

  The display 48, the keyboard 49, the mouse 50, and the RW device 51 may be omitted, and may be connected to the bus 53 as necessary.

  The storage unit 54 stores a program and information for causing the computer 42 to function as the management device 12. Specifically, the storage unit 54 stores a management program 55 and a table 62. The management program 55 includes an information collection process 56, a relay point calculation process 57, a migration management process 58, and a virtualized resource control process 59. The management program 55 can include a virtual resource management process 60. The table 62 includes a virtualization infrastructure table 63, a virtual machine table 64, a virtualization infrastructure operating rate information table 65, a virtualization resource empty / busy management table 66, and a virtualization resource timer management table 67. The storage unit 54 stores each piece of information constituting the table 62, and each piece of information constituting the table 62 is developed in the memory 46 to become a table.

  The CPU 44 reads out the management program 55 from the storage unit 54 and expands it in the memory 46, and sequentially executes the processes included in the management program 55, whereby the computer 42 operates as the management apparatus 12 shown in FIG. The CPU 44 operates as the information collection management function unit 13 shown in FIG. 1 by executing the information collection process 56, and operates as the relay point calculation function unit 14 shown in FIG. 1 by executing the relay point calculation process 57. To do. The CPU 44 operates as the migration management function unit 15 shown in FIG. 1 by executing the migration management process 58. Further, the CPU 44 operates as the virtual resource control function unit 17 illustrated in FIG. 1 by executing the virtual resource control process 59. When the management program 55 includes the virtual resource management process 60, the CPU 44 operates as the virtual resource management function unit 18 illustrated in FIG. 1 by executing the virtual resource management process 60.

  When the management device 12 is realized by the computer 42, the storage area of the storage unit 54 that stores each piece of information in the table 62 is used as the storage unit 16 that stores the table shown in FIG. That is, the storage area of each information constituting the table 62 stored in the storage unit 54 corresponds to the storage area of the storage unit 16 storing each information of the table shown in FIG. Each information stored in the storage area is expanded to form a table.

  When the data center 22 is realized by the physical machine 70, the physical machine 70 includes a CPU 72, a memory 74, and a nonvolatile storage unit 82. The CPU 72, the memory 74, and the storage unit 82 are connected to each other via a bus 81. The physical machine 70 includes a display 76 that is a display device, and a keyboard 77 and a mouse 78 that are input devices. The display 76, the keyboard 77, and the mouse 78 are connected to a bus 81. Further, the physical machine 70 includes a device (RW device) 79 for reading and writing from / to the recording medium 69, and the RW device 79 is connected to the bus 81. The physical machine 70 includes a communication unit (I / O) 80 including an interface for connecting to the network 20. The storage unit 82 can be realized by an HDD, a flash memory, or the like.

  The display 76, the keyboard 77, the mouse 78, and the RW device 79 may be omitted, and may be connected to the bus 81 as necessary.

  The storage unit 82 stores a program and information for causing the physical machine 70 to function as the data center 22. Specifically, the storage unit 82 stores a GW program 83, a virtualization infrastructure program 84, and a virtual machine program 88. The virtualization infrastructure program 84 includes a virtualization infrastructure main process 85, a virtualization resource management process 86, and a virtualization resource control process 87.

  The CPU 72 of the physical machine 70 reads out the program from the storage unit 82, expands it in the memory 74, and executes it sequentially, whereby the physical machine 70 operates as the data center 22 shown in FIG. Specifically, the CPU 72 reads the GW program 83 from the storage unit 82, expands it in the memory 74, and executes the GW program 83, whereby the physical machine 70 operates as the GW 24 of the data center 22 shown in FIG. In addition, the CPU 72 reads the virtualization infrastructure program 84 from the storage unit 82, expands it in the memory 74, and sequentially executes the processes included in the virtualization infrastructure program 84, so that the physical machine 70 can execute the data center 22 shown in FIG. It operates as the virtualization platform 26 of the system. Furthermore, the CPU 72 operates as the virtual machine 28 illustrated in FIG. 1 by executing the virtual machine program 88.

  The CPU 72 operates as the virtualization infrastructure main function unit 32 shown in FIG. 1 by executing the virtualization infrastructure main process 85 of the virtualization infrastructure program 84. Further, the CPU 72 operates as the virtual resource management function unit 34 by executing the virtual resource management process 86, and operates as the virtual resource control function unit 36 by executing the virtual resource control process 87. .

  Further, in the following description, when the data center 22 is realized by the physical machine 70, the symbols (DC1, DC2, DC3 shown in FIG. 11) are used when the physical machine 70 or elements included in the physical machine 70 are individually targeted. ) To distinguish. Specifically, the physical machine 70 that realizes the migration source data center 22DC1 will be described as a physical machine 70DC1. Further, the physical machine 70 that realizes the data center 22DC2 at the relay point will be described as a physical machine 70DC2. Further, the physical machine 70 that realizes the destination data center 22DC3 will be described as a physical machine 70DC3.

  The management program 55 is an example of a virtual machine management program in the disclosed technology. The management program 55 is also a program for causing the management apparatus 12 to function as a virtual machine management apparatus in the disclosed technology. An HDD that is an implementation example of the storage unit 54 is an example of a recording medium of the disclosed technology. A recording medium 69 such as an optical disk is also an example of a recording medium of the disclosed technology.

  Next, the operation of the virtual machine management system 10 according to this embodiment will be described. In the first embodiment, a case is described in which one virtual machine 28 is set as a movement target, and the data center 22DC2 at the relay point is relayed from the data center 22DC1 at the movement source to the data center 22DC3 at the movement destination. To do.

  FIG. 12 shows an example of the flow of processing in the computer 42 that operates as the management apparatus 12 when the virtual machine management system 10 is realized by the computer system 40.

  In the computer system 40, the management program 55 is executed by the computer 42, whereby the computer 42 operates as the management device 12 (FIG. 1). In addition, the physical machine 70 operates as the data center 22 by executing the GW program 83, the virtualization infrastructure program 84, and the virtual machine program 88 in the physical machine 70.

  The CPU 44 of the computer 42 that operates as the management apparatus 12 executes the information collection process 56, acquires various types of information, and registers the collected various types of information in the table 62. That is, the CPU 44 of the computer 42 operates as the information collection management function unit 13 by executing the information collection process 56 included in the management program 55, and periodically, in response to a user instruction or a process request described later. Execute information collection processing. Further, the information collection management function unit 13 registers the collected various information in the table 62.

  For example, when the virtualization infrastructure 26 is constructed in the data center 22, the information collection management function unit 13 collects information indicating the constructed virtualization infrastructure 26, and stores the collected information in the virtualization infrastructure table 63 ( (See FIG. 2). In addition, when information indicating the registered virtualization infrastructure 26 is updated, the information collection management function unit 13 updates the virtualization infrastructure table 63 with the information indicating the updated virtualization infrastructure 26.

  When the virtual machine 28 is constructed on the virtualization platform 26, the information collection management function unit 13 collects information indicating the constructed virtual machine 28, and stores the collected information in the virtual machine table 64 (see FIG. 3). ). Further, when information indicating the registered virtual machine 28 is updated, the information collection management function unit 13 updates the virtual machine table 64 with the information indicating the updated virtual machine 28. In addition, when collecting information indicating the virtual machine 28, the information collection management function unit 13 assigns a resource ID for identifying resource information indicating a resource used in the virtual machine, and the virtualization platform operation rate information table 65 (See FIG. 4). In addition, when the registered information is updated, the information collection management function unit 13 updates the virtualization infrastructure operating rate information table 65 with the updated information.

  Furthermore, the information collection management function unit 13 also collects the status of resources used in the virtual machine 28 and registers them in the virtualized resource empty / busy management table 66 (see FIG. 5). In addition, when the registered information is updated, the information collection management function unit 13 updates the virtualization infrastructure operating rate information table 65 with the updated information.

  First, in step 100, the CPU 44 of the computer 42 operating as the management apparatus 12 calculates the data center 22 of the relay point for relaying from the movement source data center 22 to the movement destination data center 22 for the movement target virtual machine 28. To do. That is, when the CPU 44 executes the relay point calculation process 57, the relay point calculation function unit 14 (FIG. 1), for example, information indicating the movement source and the movement destination of the virtual machine 28 input by the user using the keyboard 49 or the like. Depending on the above, the relay point and route (order) are calculated. For example, as the relay point and route (order), the travel time from the virtualization platform 26 of the location information indicated by “HV IP” to the virtualization platform 26 of the location information indicated by “HV IP” The relay point and route with the shortest are calculated. In other words, the virtualization platform 26 including the virtual machine 28 to be moved is the next virtualization platform 26 to receive the virtual machine 28, and the virtualization platform 26 having the shortest migration time is transferred from the migration source virtualization platform 26 to the migration destination. The virtualization platform 26 is sequentially obtained. Information indicating the calculated relay point and route (order) is stored as a relay point list 68 (FIG. 8).

  Next, in step 102, the CPU 44 acquires information about the virtual machine 28 and the destination data center 22 from the source via the relay point. That is, the virtual resource control function unit 17 (FIG. 1) acquires information on the virtual machine 28 and the data center 22 when the CPU 44 executes information acquisition processing in the virtual resource control process 59. First, the virtualization resource control function unit 17 refers to the relay destination list 68 (FIG. 8), and moves the physical machine 70 and the route (order) from the source to the destination via the relay site for the virtual machine 28 to be moved. ) Is acquired. Here, a case where the information of the route number “1” is acquired from the relay point list 68 will be described. That is, the virtualization infrastructure 26 indicated by “H0001” is the migration source, and the virtualization infrastructure 26 indicated by “H0003” is the migration destination. Further, the virtualization infrastructure 26 indicated by “H0002” and “H0004” is a relay point. In order to simplify the following description, the relay point will be described on behalf of the virtualization platform 26 indicated by “H0002”.

  The virtualized resource control function unit 17 acquires resource information used by the virtual machine 28 to be moved. Acquisition of resource information used by the migration target virtual machine 28 is acquired by a response to the instruction to acquire the resource information to the information collection management function unit 13. That is, the information collection management function unit 13 refers to the virtual machine table 64 (FIG.) 3 and the virtualization infrastructure operating rate information table 65 (FIG. 4), and responds with resource information used by the virtual machine 28 to be moved. Here, the virtualized resource control function unit 17 acquires the resource information used by the virtual machine 28 with the resource ID indicated by “VRSid001”.

  Next, in step 104, the CPU 44 executes a request for securing resources of the virtual machine 28 to be moved in the relay point and the data center 22 of the movement destination. In other words, the virtual resource control function unit 17 (FIG. 1) requests the data center 22 to secure resources by the CPU 44 executing the resource securing process in the virtualized resource control process 59. That is, the virtual resource control function unit 17 uses the resource information (information indicated by the resource ID “VRSid001”) used by the virtual machine 28 to be moved to the physical machine 70 that operates as the relay point and the data center 22 of the movement destination. A request to secure the resource indicated by is executed. An example of resource information used by the migration target virtual machine 28 includes information indicating a resource usage rate, such as a CPU load, a CPU usage rate, and a memory usage amount.

  Next, in step 106, the CPU 44 executes live migration for moving the virtual machine 28 to be moved. That is, when the CPU 44 executes the migration management process 58, the migration management function unit 15 (FIG. 1) instructs to move the virtual machine 28 to be moved. When the execution of live migration is completed, in the next step 108, the release of the resource requested to be secured in step 104 is instructed. That is, the virtual resource control function unit 17 (FIG. 1) releases (releases) the resource with the resource ID secured by the execution of step 102 by the CPU 44 executing the release process of the virtual resource control process 59. Request.

  On the other hand, various processes are executed in the physical machine 70 operating as the data center 22 in accordance with an instruction from the computer 42 operating as the management apparatus 12.

  Note that the CPU 72 of the physical machine 70 can exchange information with the management apparatus 12 or another data center 22 by executing the GW program 83. Further, the CPU 72 can construct a virtual environment by using a virtualization technology by executing the virtualization platform program 84. Furthermore, the CPU 72 can construct the virtual machine 28 in the virtual environment by executing the virtual machine program 88.

  FIG. 13 shows an example of the flow of processing by the virtualization platform program 84 executed on the physical machine 70 operating as the data center 22.

  In step 200, the CPU 72 of the physical machine 70 determines whether or not the instruction from the computer 42 that operates as the management apparatus 12 is a resource reference request for acquiring the resource information of the migration target virtual machine 28. . If the determination in step 200 is affirmative, the CPU 72 proceeds to step 202, executes the resource reference process of the migration target virtual machine 28, and ends this processing routine. On the other hand, if the determination in step 200 is negative, the CPU 72 proceeds to step 204 and determines whether or not the instruction from the computer 42 operating as the management apparatus 12 is a request for securing the resources of the virtual machine 28 to be moved. . If the determination in step 204 is affirmative, the CPU 72 proceeds to step 206, executes the resource securing process for the virtual machine 28 to be moved, and ends this processing routine.

  If the determination in step 204 is negative, the CPU 72 proceeds to step 208 and determines whether or not the instruction from the computer 42 operating as the management apparatus 12 is a movement request for the virtual machine 28 to be moved. If the determination in step 208 is affirmative, the CPU 72 proceeds to step 210, executes the movement process of the virtual machine 28 to be moved, and ends this processing routine. On the other hand, if the determination in step 208 is negative, the CPU 72 proceeds to step 212 and determines whether or not the instruction from the computer 42 operating as the management apparatus 12 is a resource release request secured in step 206. If the determination in step 212 is affirmative, the CPU 72 proceeds to step 214, executes a secured resource release process, and ends this processing routine.

  Next, in the virtual machine management system 10, securing resources in the relay point and the data center 22 at the destination before live migration will be further described.

  FIG. 14 illustrates an example of information exchange regarding resource reference and resource reservation of the virtual machine 28 in the virtual machine management system 10.

  The relay point calculation function unit 14 of the management apparatus 12 calculates a relay point and a route (order) according to the input information indicating the movement source and the movement destination of the virtual machine 28 (step 100). The virtualization resource control function unit 17 acquires information about the virtual machine 28 and the data center 22 (step 102). That is, the information of the route number “1” is acquired from the relay point list 68, the source virtualization infrastructure 26 is “H0001”, the relay destination virtualization infrastructure 26 is “H0002”, and the destination virtualization infrastructure 26 is “H0003”. Further, the migration target virtual machine 28 is the virtual machine 28 indicated by the VM Name “VM-A1-1” with the VMID “V0001”. That is, the virtualized resource control function unit 17 acquires the resource information (FIG. 3) of the virtual machine 28 indicated by the VM Name “VM-A1-1” of the VMID “V0001”. Specifically, the management apparatus 12 receives a response from the physical machine 70 of the migration source data center 22 and stores the resource information of the virtual machine 28 indicated by the migration target VMID “V0001” in the virtual machine table 64 (FIG. Register in 3). Further, the management apparatus 12 registers the resource information of the virtual machine 28 indicated by the received migration target VMID “V0001” in the virtualization infrastructure operating rate information table 65 with the resource ID “VRSid001”.

  On the other hand, the physical machine 70 of the migration source data center 22 executes resource reference processing of the migration target virtual machine 28 in response to a request from the management apparatus 12 (step 202). That is, the virtual resource management function unit 34 of the migration source data center 22 receives a request from the management apparatus 12, and refers to the resource of the virtual machine 28 indicated by the VMID “V0001” to be migrated to the virtual resource control function unit 36. Instruct. The virtual resource control function unit 36 confirms the resource of the virtual machine 28 to be moved, and sends it back to the virtual resource management function unit 34 as resource information. The virtual resource management function unit 34 responds to the management apparatus 12 with resource information indicating the resources of the migration target virtual machine 28.

  When the information acquisition is completed, the management device 12 requests the virtual resource control function unit 17 to secure the resources of the virtual machine 28 to be moved from the relay center and the data center 22 of the movement destination (step 104). That is, the virtualization resource control function unit 17 requests the virtualization infrastructure 26 of “H0002” at the relay point and the virtualization infrastructure 26 of “H0003” at the movement destination to secure the resource indicated by the resource ID “VRSid001”. .

  The physical machine 70 of the data center 22 at the relay site executes the resource securing process for the virtual machine 28 to be moved (step 206). In other words, the virtual resource management function unit 34 in the data center 22 at the relay point accepts a request for securing the resource with the resource ID “VRSid001” from the management apparatus 12, and the virtual machine 28 to be moved to the virtual resource control function unit 36. Instruct to secure resources. The virtualization resource control function unit 36 instructs the virtualization platform main function unit 32 to make a resource reservation request in accordance with the resource reservation instruction. The virtualization infrastructure main function unit 32 executes a process for securing a resource in response to a resource securing instruction from the virtualization resource control function unit 36, and obtains resource securing process result information indicating a processing result. 36 is notified. An example of the resource securing process result information is information indicating that resource securing has been completed. The virtual resource control function unit 36 sends a notification of resource reservation processing result information to the virtual resource management function unit 34. The virtual resource management function unit 34 responds to the management apparatus 12 with the resource securing process result information. Note that the resource securing process result information includes information indicating the secured resource when indicating that the securing of the resource of the migration target virtual machine 28 is completed.

  The management device 12 receives a response from the physical machine 70 of the data center 22 at the relay site, assigns a resource ID “VRSid007”, and a virtual infrastructure availability information table 65 (FIG. 4) and a virtual resource empty / busy management table 66 (FIG. 5). In the virtualization platform operation rate information table 65, the virtual machine 28 has not been determined for the information indicating the resources secured by the physical machine 70, so the information indicating the VMID, VM Name, and NW Name is not yet determined. "NONE" information indicating that is stored. Also, “temporary” information indicating the secured state is stored in the empty / busy information of the virtual resource empty / busy management table 66 corresponding to the resource ID “VRSid007” of the virtual infrastructure operating rate information table 65.

  In the present embodiment, when resources are secured in the physical machine 70, a time limit value for releasing resources secured when not in use can be determined. For example, when information of “temporary” indicating the secured state is stored in the air / busy information of the resource ID “VRSid007”, the time limit value is determined with reference to the virtualization resource timer management table 67. Specifically, the resource ID “VRSid007” is set to the resource class “VRS_class0” corresponding to the type and size of a predetermined resource used by the virtual machine 28. Since the resource class “VRS_class0” has the time limit value “T001” (see the time limit value table 67B shown in FIG. 7), the time limit value of the resource ID “VRSid007” becomes “T001” (see FIG. 6). (See VM timed timer table 67A shown).

  Similarly, the physical machine 70 of the migration destination data center 22 executes the resource securing process of the migration target virtual machine 28 (step 206).

  When the management apparatus 12 finishes securing the resources of the migration target virtual machine 28, the management apparatus 12 executes live migration for moving the migration target virtual machine 28 (step 106).

  FIG. 15 shows an example of the flow of information acquisition processing in step 102 (FIG. 12) executed by the computer 42 that operates as the management apparatus 12. FIG. 16 shows an example of the flow of resource reference processing in step 202 (FIG. 13) executed by the physical machine 70 operating as the migration source data center 22.

  In step 110, the CPU 44 of the computer 42 operating as the virtual resource control function unit 17 of the management apparatus 12 refers to the relay location list 68, and information on the data center 22 that is the destination from the source via the relay location. To get. Next, in step 112, the CPU 44 acquires resource information used by the migration target virtual machine 28 operating at the migration source. That is, the CPU 44 requests resource information from the physical machine 70 of the migration source data center 22 and acquires the resource information indicated by the resource ID “VRSid001” as a response.

  On the other hand, the CPU 72 of the physical machine 70 operating as the migration source data center 22 accepts a request from the management device 12 at the virtualization resource management function unit 34 in step 220 and sends the virtual machine 28 to the virtualization resource control function unit 36. Indicates resource reference. Next, in step 222, the CPU 72 executes an acquisition process for confirming the resources of the virtual machine 28 to be moved by the virtualization resource control function unit 36 and returns the acquisition process to the virtualization resource management function unit 34. Next, in step 224, the CPU 72 responds with resource information indicating the resource of the virtual machine 28 to be moved as an acquisition result in the virtual resource management function unit 34.

  FIG. 17 shows an example of the flow of processing for requesting resource reservation in step 104 (FIG. 12) executed by the computer 42 operating as the management apparatus 12. FIG. 18 shows an example of the flow of resource securing processing in step 206 (FIG. 13) executed by the physical machine 70 operating as the data center 22.

  In step 114, the CPU 44 of the computer 42 that operates as the virtualization resource control function unit 17 of the management apparatus 12 requests the relay center and the destination data center 22 to secure the resources of the virtual machine 28 to be moved. That is, the CPU 44 requests to secure the resource by the resource information of the resource ID “VRSid001”.

  On the other hand, the CPU 72 of the physical machine 70 in the data center 22 receives the request from the management apparatus 12 at the virtual resource management function unit 34 in step 226 and transfers the resource of the virtual machine 28 to be moved to the virtual resource control function unit 36. Instruct to secure. That is, the CPU 72 instructs to secure a resource based on the resource information of the resource ID “VRSid001”. Next, in step 228, the CPU 72 uses the virtualization resource control function unit 36 to make a request for securing resources to the virtualization infrastructure main function unit 32. In the next step 230, the CPU 72 executes a process for securing resources in the virtualization infrastructure main function unit 32, and notifies the virtualization resource control function unit 36 of resource securing process result information indicating a processing result. In addition, the virtual resource control function unit 36 sends a notification of resource securing process result information to the virtual resource management function unit 34. Next, in step 232, the CPU 72 uses the virtual resource management function unit 34 to respond to the management apparatus 12 with resource reservation processing result information for the resources of the virtual machine 28 to be moved.

  For example, consider a case in which the resources of the second virtual machine 28B are secured in the physical machine 70 of the data center 22 while the first virtual machine 28A is operating. When the CPU usage rate used by the virtual machine 28 is a resource, the CPU resource 33 uses three cells 33A when the first virtual machine 28A is operating at a CPU usage rate of 30% (for 6 GHz). (FIG. 9). When receiving a request to reserve a resource having a CPU usage rate of 20% (for 4 GHz) of the second virtual machine 28B in the physical machine 70 of the data center 22, the virtualized resource management function unit 34 starts from the free resource. Securing two cells 33B is executed.

  Also, for example, when the usage rate of the memory used by the virtual machine 28 is used as a resource, the memory resource 35 uses two cells 35A when the first virtual machine 28A is operating with a memory usage amount of 8 GB. (FIG. 10). When receiving a request to secure a resource that is the memory usage of 12 GB of the second virtual machine 28B to the physical machine 70 of the data center 22, the virtualization resource management function unit 34 secures the third cell 35B. Execute.

  By the way, in the physical machine 70 of the data center 22, securing of resources may fail. Therefore, the management apparatus 12 excludes the data center 22 that failed to secure the resource, recalculates the relay point and route (order) to the destination, and continues the processing.

  In other words, the CPU 44 of the computer 42 operating as the virtualized resource control function unit 17 of the management apparatus 12 determines in step 116 whether the response to the request in step 114 is a successful resource reservation. If the determination in step 116 is affirmative, this processing routine is terminated as it is. On the other hand, if a negative determination is made in step 116, the process proceeds to step 118. In step 118, an instruction to recalculate the data center 22 at the relay point from the movement source to the movement destination is removed except for the relay point where the resource securing has failed, and the relay point data center 22 is set to the relay point calculation function unit 14. Re-acquire the indicated information. The re-acquisition of the information indicating the relay point can be executed by selecting a route from the relay point list 68 excluding the relay point that failed to secure the resource. Next, the CPU 44 returns to step 114 and executes a resource securing process according to the information indicating the data center 22 of the relay site that has been reacquired.

  If the CPU 44 makes an affirmative determination in step 116, it can set a time limit value in order to release resources reserved in the physical machine 70 when not in use. For example, for the resource with the resource ID “VRSid007”, the resource class “VRS_class0” is set, the time limit value is “T001” (see the time limit value table 67B shown in FIG. 7), and the time limit value is “T001”. (See VM timed timer table 67A shown in FIG. 6).

  Next, live migration processing in the virtual machine management system 10 will be further described.

  FIG. 19 shows an example of information exchange about live migration in the virtual machine management system 10.

  The migration management function unit 15 of the management apparatus 12 confirms the acceptance of the virtual machine 28 to the accepting data center 22 in accordance with the order of movement, and then instructs the movement of the virtual machine 28 (step 106).

  First, the case where the virtual machine 28 is moved from the data center 22 of the movement source to the data center 22 of the relay site will be described. In this case, the data center 22 at the relay site is an accepting destination for the virtual machine 28 operating in the data center 22 at the movement source. The virtual machine 28 is moved between the data center 22 of the movement source and the data center 22 of the relay destination as the reception destination (Step 210).

  The physical machine 70 of the receiving data center 22 confirms whether or not the virtual machine 28 to be moved can be received by the virtualization infrastructure main function unit 32 in response to a request from the management apparatus 12, and the confirmation result Reply. Next, when the virtual machine 28 can be accepted at the receiving destination, the physical machine 70 of the migration source data center 22 performs the virtual machine 28 in the virtualization infrastructure main function unit 32 in accordance with an instruction from the management device 12. Execute the move process. The physical machine 70 of the receiving data center 22 executes the receiving process of the virtual machine 28 in the virtualization infrastructure main function unit 32, and returns the processing result to the moving source data center 22. The virtualization infrastructure main function unit 32 of the migration source data center 22 responds to the management apparatus 12 with information indicating the processing result of the migration process of the virtual machine 28.

  When the virtual machine 28 can be accepted at the accepting destination, the management apparatus 12 sets a value of “temporary on migration” indicating the migration status in the empty / busy information of the virtual resource empty / busy management table 66 (FIG. 5). Register and instruct the virtual machine 28 to start moving. In other words, the vacancy information of the resource ID “VRSid001” at the migration source is updated to “temporary on migration”, and the vacancy information of the resource ID “VRSid007” at the reception destination is updated from “temporary” to “temporary on migration”.

  Next, a case where a virtual machine is moved from the data center 22 at the relay point to the data center 22 as the movement destination will be described. In this case, the data center 22 at the relay site is the data center that is the migration source of the virtual machine, and the data center 22 that is the migration destination is the reception destination of the virtual machine. In the same manner as when the virtual machine 28 is moved from the data center 22 at the transfer source to the data center 22 at the relay destination as the reception destination, between the data center 22 at the relay destination and the data center 22 at the transfer destination as the reception destination. The virtual machine 28 is moved (step 210).

  FIG. 20 shows an example of the movement process in step 106 (FIG. 12) executed by the computer 42 operating as the management apparatus 12. FIG. 21 shows an example of the flow of migration processing in step 210 (FIG. 13) executed by the physical machine 70 operating as the migration source and reception destination data center 22.

  In step 120, the CPU 44 of the computer 42 operating as the migration management function unit 15 of the management apparatus 12 notifies the acceptance data center 22 of acceptance information requesting acceptance of the migration source virtual machine 28. Next, in step 122, the CPU 44 determines whether or not the reply from the receiving data center 22 is information indicating that it can be accepted. If an affirmative determination is made in step 122, the CPU 44 shifts the processing to step 124 and instructs the virtualization platform 26 in the migration source data center 22 to move the migration target virtual machine 28.

  Next, in step 126, the CPU 44 determines whether or not the reply from the migration source data center 22 is information indicating that the migration of the virtual machine 28 was successful. If the CPU 44 makes an affirmative determination in step 126, in the next step 128, it is determined whether or not there is a next data center 22 in the movement order of the virtual machine 28 by determining whether or not the data center 22 is the final migration destination. Judge whether or not. If the next data center 22 exists, the CPU 44 makes a negative determination in step 128 and returns the process to step 120. On the other hand, if the CPU 44 makes an affirmative determination in step 128, it is assumed that the movement of the virtual machine 28 has all been successful in accordance with the movement order, and this processing routine ends.

  On the other hand, the CPU 72 of the physical machine 70 that operates as the virtualization infrastructure main function unit 32 of the data center 22 determines whether or not its virtualization infrastructure 26 is the reception destination of the virtual machine 28 in step 240. The determination in step 240 can be made by determining whether or not acceptance information has been received from the management device 12. When the own virtualization platform 26 is the reception destination of the virtual machine 28, the CPU 72 makes an affirmative determination at step 240 and shifts the processing to step 242. In step 242, the CPU 72 confirms whether or not the virtual machine 28 can be accepted, and returns a confirmation result to the management apparatus 12. Next, in step 244, the CPU 72 determines whether or not the confirmation result returned to the management apparatus 12 can accept the virtual machine 28. If the determination result is negative, this processing routine is terminated. On the other hand, if the CPU makes an affirmative determination in step 244, the process proceeds to step 246, executes a process of accepting the virtual machine 28 in the virtualization platform 26, and notifies the management apparatus 12 of the process result.

  If the own virtualization infrastructure 26 is the virtualization infrastructure 26 that manages the migration target virtual machine 28, the CPU 72 makes a negative determination in step 240 and shifts the processing to step 248. In step 248, the CPU 72 executes a process of moving the virtual machine 28 to be moved to the receiving destination, and returns a processing result to the management apparatus 12 in the next step 250.

  Incidentally, it may be difficult for the physical machine 70 in the data center 22 to accept the virtual machine 28 depending on the operation status. In addition, the movement of the virtual machine 28 may fail. Therefore, when it is difficult for the management device 12 to accept the virtual machine 28 at the data center 22 or when the movement of the virtual machine 28 fails, the management device 12 excludes the corresponding data center 22 and relay points and routes ( The order is calculated again and the processing is continued.

  In other words, the CPU 44 of the computer 42 operating as the management apparatus 12 notifies the virtualization resource control function unit 17 from the migration management function unit 15 that the acceptance of the virtual machine 28 is difficult or the migration has failed in step 132. . The virtualization resource control function unit 17 instructs the relay point calculation function unit 14 to recalculate the relay point data center 22 reaching the destination, similarly to the process of step 118 shown in FIG. Information indicating the data center 22 is reacquired. In step 132, if it is difficult to accept the virtual machine 28, an instruction to recalculate the data center 22 at the relay point reaching the destination is given to the relay point calculation function unit 14. In addition, since it is considered that the process is in the process of moving the virtual machine 28, if the movement fails, in step 132, the data center 22 of the relay point reaching the destination is recalculated with the failed relay point as the movement source. An instruction is given to the relay point calculation function unit 14. In the next step 134, similarly to the processing shown in FIG. 17, the relay point and the destination resource are secured, and this processing routine ends.

  Next, in the virtual machine management system 10, the release of the resources secured in the relay point and the destination data center 22 before the movement of the virtual machine 28 will be further described.

  FIG. 22 shows an example of information exchange regarding the resource release of the virtual machine 28 in the virtual machine management system 10.

  In the management apparatus 12, when the migration process of the virtual machine 28 by the migration management function unit 15 is completed (step 106), the virtual resource control function unit 17 requests the release of the reserved virtual machine 28 resources (step 108). ). In other words, the management device 12 requests the release of the resources reserved for the movement of the virtual machine 28 based on the virtual resource empty / busy management table 66. Specifically, the management device 12 sets the resource of the reserved virtual machine 28 to a resource that has been used for a predetermined period of time after the migration process has been completed, or a predetermined period of time that has not been used and has been reserved. Request release of elapsed resources.

  In response to a request from the management apparatus 12, the physical machine 70 of the migration source data center 22 executes a secured resource release process for releasing (releasing) the resources of the secured virtual machine 28 (step 214). That is, the virtual resource management function unit 34 receives a request from the management device 12 and instructs the virtual resource control function unit 36 to release the resources of the reserved virtual machine 28. The virtualization resource control function unit 36 instructs the virtualization platform main function unit 32 to release a resource in response to a resource release instruction. The virtualization infrastructure main function unit 32 releases resources in response to a resource release instruction from the virtualization resource control function unit 36, and notifies the virtualization resource control function unit 36 of the completion of resource release when the release is completed. The virtual resource control function unit 36 sends a resource release completion notification to the virtual resource management function unit 34. The virtual resource management function unit 34 responds to the management apparatus 12 with information indicating that the release of the resources of the virtual machine 28 has been completed.

  Similarly, the physical machine 70 of the relay point and the data center 22 of the movement destination sequentially executes the resource release process of the virtual machine 28 (step 214).

  FIG. 23 shows an example of the flow of resource release request processing in step 108 (FIG. 12) executed by the computer 42 operating as the management apparatus 12. FIG. 24 shows an example of the flow of resource release processing in step 214 (FIG. 13) executed by the physical machine 70 operating as the data center 22.

  In step 140, the CPU 44 of the computer 42 that operates as the virtual resource control function unit 17 of the management apparatus 12 acquires one resource ID from the virtual resource empty / busy management table 66 (FIG. 5). Next, in step 142, the CPU 44 determines whether or not the processing for all resource IDs has been completed by determining whether or not the processing for the last resource ID of the resource ID acquired in step 140 has been completed. If the determination is affirmative in step 142, the CPU 44 ends the processing routine as it is. If the determination is negative, the CPU 44 proceeds to step 144.

  In step 144, the CPU 44 determines whether or not the resource state of the resource ID acquired in step 140 is “temporary” in the secured state. If the determination is affirmative, the process proceeds to step 146. Shifts the processing to step 152. In step 146, the CPU 44 refers to the virtual resource empty / busy management table 66, calculates the elapsed time T (time obtained by subtracting the last update time from the current time) in the secured state, and proceeds to step 148. In step 148, the CPU 44 determines whether or not the elapsed time T exceeds a time T1 that is predetermined for the secured state. The predetermined time limit value can be used as the predetermined time T1 after the resource is secured (after affirmative determination in step 116 shown in FIG. 17). For example, the resource ID “VRSid007” is set with the time limit value “T001” in the resource class “VRS_class0” (FIGS. 6 and 7). If the elapsed time T is within the time T1, the reserved resource may be used. Therefore, the CPU 44 makes a negative determination in step 148, returns the process to step 140, and executes the process for the next resource ID. . On the other hand, if the elapsed time T exceeds the time T1, the CPU 44 makes an affirmative determination in step 148, and after instructing the release of the resource in step 150, returns the process to step 140 and executes the process for the next resource ID. . In step 150, when an instruction to release a resource is given, an instruction including information indicating a resource ID is given.

  When the CPU 44 makes a negative determination in step 144, the CPU 44 determines in step 152 whether or not the resource state of the resource ID acquired in step 140 is “temporary on migration” which is a migration state. If the determination is affirmative in step 152, the CPU 44 proceeds to step 154. If the determination is negative, the CPU 44 returns the process to step 140. In step 154, the CPU 44 refers to the virtual resource empty / busy management table 66, calculates the elapsed time T (time obtained by subtracting the last update time from the current time) in the migration state, and proceeds to step 156. In step 156, the CPU 44 determines whether or not the elapsed time T exceeds a time T2 that is predetermined for the in-migration state. The predetermined time T2 may be the same as the time T1, or may be different. In the present embodiment, a case where time T1 and time T2 are the same will be described. If the elapsed time T is within the time T2, the reserved resource may be used. Therefore, the CPU 44 makes a negative determination in step 156, returns the process to step 140, and executes the process for the next resource ID. . On the other hand, when the elapsed time T exceeds the time T2, the CPU 44 makes an affirmative determination in step 156, and after instructing the release of the resource in step 158, returns the process to step 140 and executes the process for the next resource ID. . In step 158, when an instruction to release the resource is given, an instruction including information indicating the resource ID is given.

  On the other hand, the CPU 72 of the physical machine 70 in the data center 22 accepts a resource release request from the management apparatus 12 at the virtual resource management function unit 34 in step 260, and the virtual machine 28 that has been secured to the virtual resource control function unit 36. Instructs the release of resources. That is, the CPU 72 instructs the virtualization resource control function unit 36 to instruct the release of the resource designated by the information indicating the resource ID. Next, in step 262, the CPU 72 makes a request to the virtualization infrastructure main function unit 32 to execute a process for releasing the secured resource in the virtualization resource control function unit 36. In the next step 264, the CPU 72 releases the resource of the instructed resource ID in the virtualization infrastructure main function unit 32, and when the release is completed, notifies the virtualization resource control function unit 36 of the resource release completion. In addition, the virtual resource control function unit 36 sends a resource release completion notification to the virtual resource management function unit 34. Next, in step 266, the CPU 72 uses the virtualization resource management function unit 34 to respond to the management apparatus 12 with information indicating that the release of the resources of the virtual machine 28 has been completed.

  As described above, in the first embodiment, before the virtual machine 28 is moved, for example, live migration is performed, the resources used by the virtual machine 28 to be moved are secured in the relay site and the destination environment. Execute live migration. That is, before moving the virtual machine 28, the resources used by the virtual machine 28 can be secured in advance, and the virtual machine 28 can be moved reliably. Further, since the resource used by the virtual machine 28 is secured in advance, the virtual machine 28 is moved before the virtual machine 28 is moved, so that the performance of the virtual machine 28 is not deteriorated due to insufficient resources.

  In the first embodiment, the relay point and route (order) to be relayed from the source to the destination are calculated by the relay point calculation function unit 14, and based on the information indicating the calculated relay point and route (order). The virtual machine 28 is moved. Therefore, a relay point and a part for calculating a route (order) for moving the virtual machine 28 can be separated as a configuration, and the virtual machine 28 from the movement source to the movement destination using the information calculated by the relay point calculation function unit 14. Can be reliably moved.

  Furthermore, in the first embodiment, when it is difficult to secure the resources used by the virtual machine 28, the relay point and the route (order) relayed from the movement source to the movement destination are reacquired. Therefore, before the movement of the virtual machine 28, it is possible to specify a relay point and a route that can secure the resources used by the virtual machine 28, and it is possible to stop the movement of the virtual machine 28 including the possibility of performance degradation. Since the virtual machine 28 is moved according to the relay point and route specified by the reacquisition, the virtual machine 28 can be moved without degrading the performance of the virtual machine 28 to be moved.

  In the first embodiment, when the movement of the virtual machine 28 becomes difficult, the relay point and the route relayed from the movement source that has become difficult to move to the movement destination are reacquired. . Therefore, even when the virtual machine 28 is moving, it is possible to specify a relay point and a route that can secure the resources used by the virtual machine 28, and the virtual machine 28 can be moved without degrading the performance of the virtual machine 28 to be moved. it can.

  Furthermore, in the first embodiment, after a resource used by the virtual machine 28 to be moved is secured, the secured resource is released when the resource is unused for a predetermined time. Therefore, it is possible to suppress maintaining resources in a state in which the resources are secured for a long time. In the first embodiment, the reserved resources are released when a predetermined time elapses after the virtual machine 28 to be moved is moved. Therefore, it is possible to prevent the resources from being maintained in a state in which the virtual machines 28 are secured for a long time after the movement.

Second Embodiment
Next, a second embodiment will be described. In the second embodiment, the disclosed technology is applied when moving a plurality of virtual machines. Since the second embodiment has substantially the same configuration as the first embodiment, the same parts are denoted by the same reference numerals and detailed description thereof is omitted.

  FIG. 25 illustrates an example of information exchange regarding resource reference and resource reservation of the virtual machine 28 in the virtual machine management system 10 according to the second embodiment.

  The relay point calculation function unit 14 of the management apparatus 12 calculates a relay point and a route (order) (step 100), and the virtualization resource control function unit 17 acquires information of the virtual machine 28 and the data center 22 (step). 102). In the second embodiment, the virtual machine 28 to be moved is a virtual machine 28 whose VM IDs are indicated by “V0001”, “V0002”, and “V0003”. The migration source data center 22 executes the resource reference process of the migration target virtual machine 28 indicated by the VM IDs “V0001”, “V0002”, and “V0003” (step 202). The management apparatus 12 assigns resource IDs “VRSid001”, “VRSid002”, and “VRSid003” to each resource information of the plurality of virtual machines 28 and registers them in the virtualization infrastructure operating rate information table 65, and relay points and destinations Is requested to secure the resources of the virtual machine 28 (step 104).

  The data center 22 at the relay site executes a resource securing process for the virtual machine 28 to be moved (step 206). Although details will be described later, in the second embodiment, it is not required to secure each resource of the plurality of migration target virtual machines 28 in the data center 22 at the relay point. That is, in the second embodiment, on the premise that the plurality of migration target virtual machines 28 are sequentially moved, a resource area is secured so that the largest resource among the plurality of migration target virtual machines 28 can be used, Use the reserved resource area. Therefore, the management apparatus 12 instructs the data center 22 at the relay site to ensure that the maximum resource among the plurality of virtual machines 28 to be moved can be used. Then, the management device 12 stores “temporary” information indicating the secured state in the empty / busy information of the virtual resource empty / busy management table 66. On the other hand, the migration destination data center 22 executes a resource securing process for reporting each resource of the plurality of migration target virtual machines 28 (step 206).

  FIG. 26 shows an example of the flow of information acquisition processing in the management apparatus 12 according to the second embodiment. FIG. 27 shows an example of the flow of resource reference processing in the migration source data center 22 according to the second embodiment.

  The CPU 44 of the computer 42 refers to the relay location list 68 and acquires information about the destination data center 22 from the source via the relay location (step 110). Next, in step 112A, the CPU 44 acquires each resource information (resource information indicated by resource IDs “VRSid001”, “VRSid002”, and “VRSid003”) used by the plurality of migration target virtual machines 28 operating at the migration source.

  On the other hand, the CPU 72 of the physical machine 70 instructs the virtualization resource control function unit 36 to refer to the resources of the plurality of virtual machines 28 to be moved in step 220A, and checks the resources of the plurality of virtual machines 28 in step 222A. . Next, in step 224A, the CPU 72 responds with resource information indicating the resources of the plurality of virtual machines 28 to be moved as an acquisition result.

  FIG. 28 shows an example of the flow of resource securing request processing in the management apparatus 12 according to the second embodiment. FIG. 29 shows an example of the flow of resource securing processing in the data center 22 according to the second embodiment.

  In step 114A, the CPU 44 of the computer 42 that operates as the virtualized resource control function unit 17 of the management apparatus 12 sets resource information indicating the maximum resource among the resource information indicating the resources of the plurality of virtual machines 28 to be moved. Identify. Next, in step 114B, the CPU 44 requests the relay center data center 22 to secure resource information indicating the maximum resource obtained in step 114A. In step 114B, in order to make a request for securing resource information indicating the maximum resource obtained in step 114A, resource information indicating a resource greater than or equal to the maximum resource may be requested. In the next step 114C, the CPU 44 requests the migration destination data center 22 to secure each resource of the plurality of virtual machines 28 to be migrated. That is, the CPU 44 requests to secure a resource by using resource information of resource IDs “VRSid001”, “VRSid002”, and “VRSid003”.

  On the other hand, in the data center 22, the CPU 72 first determines in step 225 whether or not the request from the management apparatus 12 is for the relay point. When the request from the management apparatus 12 is for the relay point, the CPU 72 makes an affirmative determination in step 225, and in step 225A, similarly to the processing shown in FIG. 18, the resource based on the resource information indicating the requested (maximum) resource is obtained. Execute the securing process. On the other hand, when the CPU 72 makes a negative determination in step 225, the CPU 72 executes a process of securing a plurality of resources as the data center 22 of the movement destination. That is, the CPU 72 instructs the virtualization resource control function unit 36 to secure the resources of the plurality of virtual machines 28 to be moved in step 226A. Next, in step 228A, the CPU 72 uses the virtualization resource control function unit 36 to make a request for securing each resource to the virtualization infrastructure main function unit 32. In the next step 230A, the CPU 72 reserves a plurality of resources in the virtualization infrastructure main function unit 32, and notifies the virtualization resource control function unit 36 of the completion of securing a plurality of resources when the reservation is completed. In addition, the virtual resource control function unit 36 sends a plurality of resource reservation completion notifications to the virtual resource management function unit 34. Next, in step 232A, the CPU 72 responds to the management apparatus 12 with information indicating that the virtual resource management function unit 34 has secured the resources of the plurality of virtual machines 28 to be moved.

  FIG. 30 shows an example of information exchange about live migration in the virtual machine management system 10 according to the second embodiment.

  The migration management function unit 15 of the management apparatus 12 instructs the movement of the virtual machine 28 after confirming the acceptance of the virtual machine 28 to the receiving data center 22 in accordance with the movement order (step 106). Each of the plurality of virtual machines 28 is sequentially executed.

  First, when the virtual machine 28 is moved from the movement source to the relay destination as the reception destination, the virtual machine 28 is moved between the data center 22 of the movement source and the data center 22 of the relay destination as the reception destination (step) 210). In the data center 22 at the relay site, one common use resource (resource area secured as the CPU resource 33 and the memory resource 35) in which the maximum resource among the plurality of virtual machines 28 to be moved is secured is used. .

  Note that the management device 12 sets the “temporary on migration” value indicating the migration state only during the movement of each of the plurality of virtual machines 28 to the vacancy information of the virtual resource vacancy management table 66 (FIG. 5). Registration may be performed, and a value of “temporary on migration” indicating a migration in progress state may be maintained while a plurality of virtual machines 28 are moving.

  Next, when a virtual machine is moved from the data center 22 at the relay site to the data center 22 at the transfer destination, the virtual machine 28 is connected between the data center 22 at the relay location and the data center 22 as the transfer destination. It is moved (step 210). In the data center 22 at the relay point, the virtual machine is moved from the data center 22 at the relay point to the data center 22 at the transfer destination using the resources secured as the common use resource.

  FIG. 31 shows an example of the flow of migration processing in step 210 (FIG. 13) executed by the physical machine 70 operating as the migration source and reception destination data center 22 according to the second embodiment. Note that the movement process of the plurality of virtual machines 28 to be moved executed by the management apparatus 12 repeatedly executes the movement process shown in FIG. 20 (step 106 shown in FIG. 12), and thus detailed description thereof is omitted.

  When the CPU 72 of the physical machine 70 that operates as the virtualization infrastructure main function unit 32 of the data center 22 can accept the virtual machine 28 (when an affirmative determination is made at step 244), the process proceeds to step 245. In step 245, the CPU 72 determines whether or not its own virtualization infrastructure 26 is the virtualization infrastructure 26 at the relay site. The determination in step 245 can be determined by determining whether or not the received information indicates a relay point from the management device 12. If the CPU 72 makes an affirmative determination in step 245, the process proceeds to step 247, and the virtual machine uses one common usage resource in which the maximum resource among the plurality of virtual machines 28 to be moved is secured. Accept 28. On the other hand, if the CPU 72 makes a negative determination in step 245, the CPU 72 executes a process of accepting the virtual machine 28 in the virtualization platform 26 as in the process of step 246 shown in FIG. 21.

  As described above, in the second embodiment, when a plurality of virtual machines 28 are moved, the resource that is the largest of the resources used by each of the plurality of virtual machines 28 to be moved in the environment at the relay point is greater than or equal to the maximum resource. Secure one resource and execute the movement. Accordingly, when moving each of the plurality of virtual machines 28 in the data center 22 at the relay site, it is not necessary to secure the total of the resources of the plurality of virtual machines 28. As a result, in the second embodiment, in addition to the effects of the first embodiment, it is possible to suppress resource consumption due to resource reservation when moving a plurality of virtual machines 28 to be moved in an environment at a relay point. .

<Third Embodiment>
Next, a third embodiment will be described. In addition, since 3rd Embodiment is the structure similar to 1st Embodiment, it attaches | subjects the same code | symbol to the same part, and abbreviate | omits detailed description.

  In the first embodiment and the second embodiment, the resources used by the migration target virtual machine 28 are secured in the receiving data center 22 by referring to the resources used by the migration target virtual machine 28. In the third embodiment, a predetermined amount of resources are reserved in the virtualization platform 26 at a predetermined time, and when moving the virtual machine 28, the reserved resources are used as resources for securing. It is what is used. In other words, a predetermined amount of resources are reserved as resources for moving the virtual machine 28 in the data center 22 that is connected to the network 20 and has the possibility of receiving the movement of the virtual machine 28.

  FIG. 32 shows an example of information exchange until resource reference and resource reservation of the virtual machine 28 in the virtual machine management system 10.

  First, in the management apparatus 12, before the live migration process, the virtual resource control function unit 17 reserves resources determined for the resources used by the virtual machine 28 in each of the data centers 22 connected to the network 20. (Step 101). In other words, the management device 12 reserves a predetermined amount of resources as resources used by the virtual machine 28 in each of the data centers 22 by the virtualization resource control function unit 17. Specifically, the management device 12 refers to the virtualization infrastructure table 63 and requests to reserve a predetermined resource to the virtualization infrastructure 26 capable of constructing the virtual machine 28 so that it can be used when the virtual machine 28 is moved. I do. Note that the resource determined for the resource used by the virtual machine 28 can be determined by a predetermined number (for example, three times) of the resource used by the predetermined virtual machine 28.

  In each of the data centers 22, resource reservation processing from the management device 12 is executed (step 207). That is, the virtual resource management function unit 34 receives a resource reservation request from the management apparatus 12 and instructs the virtual resource control function unit 36 to make a resource reservation. The virtualization resource control function unit 36 instructs the virtualization platform main function unit 32 to make a resource reservation request in response to a resource reservation instruction. The virtualization infrastructure main function unit 32 reserves resources according to the resource reservation instruction from the virtualization resource control function unit 36, and notifies the virtualization resource control function unit 36 of completion of resource reservation when the reservation is completed. . The virtual resource control function unit 36 sends a notification of resource reservation completion to the virtual resource management function unit 34. The virtual resource management function unit 34 responds to the management apparatus 12 with information indicating that the resource reservation has been completed.

  The management device 12 receives a response from each of the data centers 22 and registers it in the virtualization infrastructure operating rate information table 65 (FIG. 4) and the virtual resource empty / busy management table 66 (FIG. 5). In the virtualization platform operation rate information table 65, since the virtual machine 28 is undetermined with respect to the information indicating the reserved resource, the information indicating the VMID, VM Name, and NW Name is undetermined. NONE "information is stored. In addition, in the vacancy information of the virtual resource vacancy management table 66, information on “temporary” indicating the secured state or “temporary on migration” indicating the in-migration state is stored. Note that dedicated information may be stored as information indicating the reserved resource in the empty / busy information of the virtualized resource empty / busy management table 66.

  FIG. 33 shows an example of the virtual infrastructure operating rate information table 65 according to the third embodiment. FIG. 34 shows an example of the virtual resource empty / busy management table 66 according to the third embodiment.

  FIG. 33 shows that resources with resource IDs “VRS002”, “VRS003”, “VRS006”, and “VRS007” are reserved in the virtualization infrastructure operating rate information table 65. FIG. 34 shows that “temporary on migration” information indicating that the resources with the resource IDs “VRS002”, “VRS003”, “VRS006”, and “VRS007” are being reserved is stored.

  As described above, in the third embodiment, since a certain amount of resources are reserved in advance before the movement of the virtual machine 28, it is possible to suppress failure in securing resources used when the virtual machine 28 is moved.

  In the above description, an example in which the virtual machine management system 10 is realized by the computer system 40 has been described. However, the present invention is not limited to these configurations, and various improvements and modifications may be made without departing from the gist described above.

  Moreover, although the aspect by which the program was previously memorize | stored (installed) in the memory | storage part was demonstrated above, it is not limited to this. For example, the management program and the virtualization platform program in the disclosed technology can be provided in a form recorded on a recording medium such as a CD-ROM or a DVD-ROM.

  All documents, patent applications and technical standards mentioned in this specification are to the same extent as if each individual document, patent application and technical standard were specifically and individually stated to be incorporated by reference. Incorporated by reference in the book.

  Regarding the above embodiment, the following additional notes are disclosed.

(Appendix 1)
When the migration target virtual machine operating on the migration source first physical machine is relayed by the second physical machine at the relay location and moved to the third physical machine at the migration destination, the first virtual machine is moved. Resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the physical machine, and positional information of each of the first physical machine, the second physical machine, and the third physical machine; An acquisition unit that acquires movement order information of a physical machine to which a virtual machine to be moved is moved;
Indicates that the resource allocation amount indicated by the resource allocation information of the migration target virtual machine is secured in each of the second physical machine and the third physical machine based on the location information acquired by the acquisition unit. A request part for transmitting request information;
From each of the second physical machine and the third physical machine, a response of the request information transmitted by the request unit is received, and based on the position information acquired by the acquisition unit, the movement order information, and the received response, An instruction unit for instructing to move a virtual machine to be moved according to the movement order information;
A virtual machine management device.

(Appendix 2)
The acquisition unit is configured to input the information indicating the migration target virtual machine operating on the migration source first physical machine, the information indicating the first physical machine, and the migration destination third physical machine. Based on the indicated information, the position information and the movement are obtained from an order calculation unit that obtains information indicating the second physical machine at the relay point and obtains movement order information of the physical machine to which the virtual machine to be moved is moved. The virtual machine management device according to appendix 1, including obtaining order information.

(Appendix 3)
When the instruction unit receives a response indicating that the resource allocation amount indicated by the resource allocation information is secured from each of the second physical machine and the third physical machine, the instruction unit selects the virtual machine to be moved. If a response indicating that the resource allocation amount has been secured has not been received from at least one of the second physical machine and the third physical machine, a second instruction that has not received the response is issued. The virtual machine management device according to Supplementary Note 1 or Supplementary Note 2, which includes issuing an instruction to reacquire physical machine position information and movement order information that are different from the physical machine.

(Appendix 4)
When the instruction unit receives information indicating that the movement of the virtual machine to be moved has failed from the second physical machine or the third physical machine, the instruction unit differs from the second physical machine. The virtual machine management device according to any one of appendix 1 to appendix 3, including an instruction to reacquire machine location information as location information of the second physical machine at the relay point.

(Appendix 5)
The acquisition unit acquires resource allocation information for each of a plurality of migration target virtual machines operating on the first physical machine,
The request unit transmits request information indicating that a resource allocation amount indicated by resource allocation information indicating that a resource allocation amount to be secured is a predetermined amount or more among the resource allocation information acquired by the acquisition unit is secured. The virtual machine management device according to any one of appendix 1 to appendix 4, including the virtual machine management device.

(Appendix 6)
Each of the plurality of physical machines further includes a securing unit that executes processing for securing a predetermined resource allocation amount,
The request unit includes transmitting request information indicating that the resource allocation amount indicated by the resource allocation information of the migration target virtual machine is secured using the resource allocation amount secured by the securing unit. The virtual machine management device according to any one of appendix 5.

(Appendix 7)
When information indicating that the virtual machine to be moved has been moved is received, the allocated resource allocation amount is allocated to each of the second physical machine and the third physical machine. The virtual machine management device according to any one of supplementary notes 1 to 6, further comprising a release unit that executes a release process.

(Appendix 8)
For at least one of the second physical machine and the third physical machine that has transmitted the request information from the request unit, even if a predetermined time elapses after the request information has been transmitted, When the information indicating that the virtual machine has been moved cannot be received, the resource allocation amount of the secured resource allocation amount is assigned to at least one of the corresponding second physical machine and the third physical machine. The virtual machine management device according to any one of supplementary notes 1 to 6, further comprising: a release unit that executes a process for releasing the allocation.

(Appendix 9)
A first physical machine that is a migration source on which the virtual machine to be moved operates, a second physical machine that relays the movement of the virtual machine that is the movement target, and a first destination that is the movement destination of the virtual machine that is the movement target A virtual machine system including three physical machines,
A computer that executes control for relaying the virtual machine to be moved from the first physical machine that is the movement source to the second physical machine that is the relay destination and moving to the third physical machine that is the movement destination.
The computer is
The resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the first physical machine is acquired, and each of the first physical machine, the second physical machine, and the third physical machine is acquired. Obtain location information and movement order information of the physical machine to which the virtual machine to be moved is moved,
Based on the acquired position information, request information indicating that the resource allocation amount indicated by the resource allocation information of the virtual machine to be moved is secured to each of the second physical machine and the third physical machine. The received request information response is received from each of the second physical machine and the third physical machine, and based on the acquired position information, movement order information, and received response, the virtual machine to be moved is Instructing movement according to the movement order information,
A virtual machine management system that executes processing.

(Appendix 10)
In a virtual machine system having a plurality of physical machines on which the virtual machine can operate, the migration target virtual machine running on the migration source first physical machine is relayed by the second physical machine at the relay site and moved. A virtual machine management method for controlling movement to a third physical machine,
On the computer,
When the computer relays the virtual machine to be moved running on the first physical machine of the movement source to the second physical machine of the relay destination and moves to the third physical machine of the movement destination, The resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the first physical machine is acquired, and each of the first physical machine, the second physical machine, and the third physical machine is acquired. Obtain location information and movement order information of the physical machine to which the virtual machine to be moved is moved,
Based on the acquired position information, request information indicating that the resource allocation amount indicated by the resource allocation information of the virtual machine to be moved is secured to each of the second physical machine and the third physical machine. And
A response of the transmitted request information is received from each of the second physical machine and the third physical machine, and based on the acquired position information, movement order information, and the received response, the virtual machine to be moved is Instruct to move according to the movement order information,
Virtual machine management method for executing processing including the above.

(Appendix 11)
In a virtual machine system having a plurality of physical machines on which the virtual machine can operate, the migration target virtual machine running on the migration source first physical machine is relayed by the second physical machine at the relay site and moved. A virtual machine management program for causing a computer to execute control to move to the third physical machine,
In the computer,
When the computer relays the virtual machine to be moved running on the first physical machine of the movement source to the second physical machine of the relay destination and moves to the third physical machine of the movement destination, The resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the first physical machine is acquired, and each of the first physical machine, the second physical machine, and the third physical machine is acquired. Obtain location information and movement order information of the physical machine to which the virtual machine to be moved is moved,
Based on the acquired position information, request information indicating that the resource allocation amount indicated by the resource allocation information of the virtual machine to be moved is secured to each of the second physical machine and the third physical machine. And
A response of the transmitted request information is received from each of the second physical machine and the third physical machine, and based on the acquired position information, movement order information, and the received response, the virtual machine to be moved is Instruct to move according to the movement order information,
Virtual machine management program that executes processing.

DESCRIPTION OF SYMBOLS 10 Virtual machine management system 12 Management apparatus 13 Information collection management function part 14 Relay point calculation function part 15 Migration management function part 16 Storage part 17 Virtualization resource control function part 18 Virtualization resource management function part 20 Network 22 Data center 24 Gateway 26 Virtualization platform 28 Virtual machine 30 GW / router function unit 32 Virtualization platform main function unit 34 Virtualization resource management function unit 36 Virtualization resource control function unit 38 Virtual machine function unit 40 Computer system 42 Computer 44 CPU
46 Memory 51 RW device 54 Storage unit 55 Management program 56 Information collection process 57 Relay point calculation process 58 Migration management process 59 Virtualization resource control process 60 Virtualization resource management process 62 Table 63 Virtualization infrastructure table 64 Virtual machine table 65 Virtualization Base operating rate information table 66 Virtualized resource idle / busy management table 67 Virtualized resource timer management table 68 Relay destination list 69 Recording medium 70 Physical machine 72 CPU
74 Memory 79 RW device 82 Storage unit 83 GW program 84 Virtualization infrastructure program 85 Virtualization infrastructure main process 86 Virtualization resource management process 87 Virtualization resource control process 88 Virtual machine program

Claims (8)

When the migration target virtual machine operating on the migration source first physical machine is relayed by the second physical machine at the relay location and moved to the third physical machine at the migration destination, the first virtual machine is moved. Resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the physical machine, and positional information of each of the first physical machine, the second physical machine, and the third physical machine; An acquisition unit that acquires movement order information of a physical machine to which a virtual machine to be moved is moved;
Indicates that the resource allocation amount indicated by the resource allocation information of the migration target virtual machine is secured in each of the second physical machine and the third physical machine based on the location information acquired by the acquisition unit. A request part for transmitting request information;
From each of the second physical machine and the third physical machine, a response of the request information transmitted by the request unit is received, and based on the position information acquired by the acquisition unit, the movement order information, and the received response, An instruction unit for instructing to move a virtual machine to be moved according to the movement order information;
A virtual machine management device. When the instruction unit receives a response indicating that the resource allocation amount indicated by the resource allocation information is secured from each of the second physical machine and the third physical machine, the instruction unit selects the virtual machine to be moved. If a response indicating that the resource allocation amount has been secured has not been received from at least one of the second physical machine and the third physical machine, a second instruction that has not received the response is issued. The virtual machine management apparatus according to claim 1, wherein an instruction to reacquire physical machine position information and the movement order information different from the physical machine is issued. The acquisition unit acquires resource allocation information for each of a plurality of migration target virtual machines operating on the first physical machine,
The request unit transmits request information indicating that a resource allocation amount indicated by resource allocation information indicating that a resource allocation amount to be secured is a predetermined amount or more among the resource allocation information acquired by the acquisition unit is secured. The virtual machine management apparatus according to claim 1 or 2. Each of the plurality of physical machines further includes a securing unit that executes a process of securing a predetermined resource allocation amount,
The request unit transmits request information indicating that the resource allocation amount indicated by the resource allocation information of the migration target virtual machine is secured using the resource allocation amount secured by the securing unit. 4. The virtual machine management device according to any one of 3 above. When information indicating that the virtual machine to be moved has been moved is received, the allocated resource allocation amount is allocated to each of the second physical machine and the third physical machine. The virtual machine management device according to claim 1, further comprising: a release unit that executes a release process. A first physical machine that is a migration source on which the virtual machine to be moved operates, a second physical machine that relays the movement of the virtual machine that is the movement target, and a first destination that is the movement destination of the virtual machine that is the movement target A virtual machine system including three physical machines,
A computer that executes control for relaying the virtual machine to be moved from the first physical machine that is the movement source to the second physical machine that is the relay destination and moving to the third physical machine that is the movement destination.
The computer is
The resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the first physical machine is acquired, and each of the first physical machine, the second physical machine, and the third physical machine is acquired. Obtain location information and movement order information of the physical machine to which the virtual machine to be moved is moved,
Based on the acquired position information, request information indicating that the resource allocation amount indicated by the resource allocation information of the virtual machine to be moved is secured to each of the second physical machine and the third physical machine. The received request information response is received from each of the second physical machine and the third physical machine, and based on the acquired position information, movement order information, and received response, the virtual machine to be moved is Instructing movement according to the movement order information,
A virtual machine management system that executes processing. In a virtual machine system having a plurality of physical machines on which the virtual machine can operate, the migration target virtual machine running on the migration source first physical machine is relayed by the second physical machine at the relay site and moved. A virtual machine management method for controlling movement to a third physical machine,
On the computer,
When the computer relays the virtual machine to be moved running on the first physical machine of the movement source to the second physical machine of the relay destination and moves to the third physical machine of the movement destination, The resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the first physical machine is acquired, and each of the first physical machine, the second physical machine, and the third physical machine is acquired. Obtain location information and movement order information of the physical machine to which the virtual machine to be moved is moved,
Based on the acquired position information, request information indicating that the resource allocation amount indicated by the resource allocation information of the virtual machine to be moved is secured to each of the second physical machine and the third physical machine. And
A response of the transmitted request information is received from each of the second physical machine and the third physical machine, and based on the acquired position information, movement order information, and the received response, the virtual machine to be moved is Instruct to move according to the movement order information,
Virtual machine management method for executing processing including the above. In a virtual machine system having a plurality of physical machines on which the virtual machine can operate, the migration target virtual machine running on the migration source first physical machine is relayed by the second physical machine at the relay site and moved. A virtual machine management program for causing a computer to execute control to move to the third physical machine,
In the computer,
When the computer relays the virtual machine to be moved running on the first physical machine of the movement source to the second physical machine of the relay destination and moves to the third physical machine of the movement destination, The resource allocation information including information on the resource allocation amount of the virtual machine to be moved in the first physical machine is acquired, and each of the first physical machine, the second physical machine, and the third physical machine is acquired. Obtain location information and movement order information of the physical machine to which the virtual machine to be moved is moved,
Based on the acquired position information, request information indicating that the resource allocation amount indicated by the resource allocation information of the virtual machine to be moved is secured to each of the second physical machine and the third physical machine. And
A response of the transmitted request information is received from each of the second physical machine and the third physical machine, and based on the acquired position information, movement order information, and the received response, the virtual machine to be moved is Instruct to move according to the movement order information,
Virtual machine management program that executes processing.

Images