JPWO2017216891A1 - Computer system, computer system control method, and program - Google Patents

Abstract

A computer system including a plurality of computers, wherein the plurality of computers includes a plurality of business computers that operate hypervisors that manage VMs, and a placement management computer that manages the placement of virtual computers that execute applications. Is a predicted value of the amount of computer resources allocated to the target application of the SLA from the hypervisor in which the VM that executes the target application of the SLA is arranged based on the process priority of the process executed on the business computer Based on the allocated resource amount, a resource amount calculation unit that calculates the allocated resource amount, and selects a hypervisor that is a placement destination of the VM that executes the SLA target application, and executes the SLA target application on the selected hypervisor. Placement instruction that instructs the placement of the VM to be executed And a part.

Description

  The present invention relates to a method for controlling a virtual machine that executes an application.

  Along with commoditization of hypervisors, various hypervisors can be selected as placement destinations of VMs that execute applications. Currently, various methods for changing the arrangement of VMs are provided. For example, a technique for arranging a VM in consideration of the amount of computer resources managed by a hypervisor is known (see, for example, Patent Document 1).

  Japanese Patent Laid-Open No. 2004-228561 collects information indicating the amount of physical resources used by each VM operating on each of the servers by the arrangement control unit of the management manager connected to the server on which the VM operates. Based on the physical resource amount used by the VM, a reservation amount, which is a physical resource amount that is expected to be further required for the operation of each VM, is calculated. Among them, the usage amount of the physical resource used by the migration target VM and the physical resource amount corresponding to the reservation amount of the VM are available, and the reservation amount of the migration target VM and the server operate. It is described that the server having the closest VM reservation amount is selected as the migration destination server, and the VM control unit 12 moves the migration target VM to the migration destination server.

JP 2011-198332 A

  In the conventional VM placement method as described in Patent Document 1, a hypervisor that is a VM placement destination is selected based on the availability of the computer resource amount. A computer resource amount greater than the computer resource amount managed by the hypervisor can be allocated to the VM.

  However, since the computer resource amount allocated to the VM depends on the computer resource amount managed by the hypervisor, the computer resource amount actually allocated to the application may be smaller than the computer resource amount set in the VM. In this case, since the performance of the application executed on the VM is deteriorated, SLA (service quality assurance) cannot be satisfied.

  Therefore, it is necessary to control the placement of VMs that execute applications in consideration of not only computer resource amounts but also SLA. An object of the present invention is to provide a system and method for controlling the placement of VMs that execute applications in consideration of SLA.

  A typical example of the invention disclosed in the present application is as follows. That is, a computer system including a plurality of computers, wherein the plurality of computers includes a processor, a memory connected to the processor, and an interface connected to the processor, and the plurality of computers manage virtual computers. A plurality of business computers on which the virtualization control unit operates, a placement management computer that manages the placement of the virtual computer that executes an application, and a virtualization control unit management computer that controls the virtualization control unit, the placement management The computer, based on the process priority of the process executed on the business computer, receives the service quality assurance target from the virtualization control unit in which the virtual computer executing the service quality assurance target application is arranged. Allocated resource amount that is a predicted value of the computer resource amount allocated to the application Based on the resource amount calculation unit to be calculated and the allocated resource amount, the virtualization control unit to be a placement destination of a virtual machine that executes the application of the service quality assurance target is selected, and the selected virtualization control A placement instruction execution unit for instructing placement of a virtual machine that executes the application subject to service quality assurance, and the virtualization control unit management computer is based on an instruction from the placement instruction execution unit. And a virtualization control unit for instructing the selected virtualization control unit to set a virtual machine that executes the application for which the service quality is to be guaranteed.

  According to the present invention, the placement management computer can control the placement of a virtual machine (VM) that executes an application in consideration of the SLA. Problems, configurations, and effects other than those described above will become apparent from the following description of embodiments.

1 is a diagram illustrating a configuration example of a computer system according to a first embodiment. FIG. 3 is a diagram illustrating an example of a hardware configuration of a computer included in the computer system according to the first embodiment. It is a figure which shows an example of the server structure information of Example 1. FIG. 6 is a diagram illustrating an example of VM process priority information according to the first embodiment. It is a figure which shows an example of AP process priority information of Example 1. FIG. FIG. 11 is a sequence diagram illustrating a flow of instance arrangement processing in the computer system according to the first embodiment. 6 is a diagram illustrating an example of an input screen according to Embodiment 1. FIG. 6 is a flowchart illustrating an example of an arrangement destination determination process executed by the arrangement management apparatus according to the first embodiment. 10 is a flowchart illustrating an example of an arrangement destination selection process executed by the arrangement management apparatus according to the first embodiment. 6 is a flowchart illustrating an example of a resource amount calculation process executed by the arrangement management apparatus according to the first embodiment. 6 is a flowchart illustrating an example of a resource amount calculation process executed by the arrangement management apparatus according to the first embodiment. 6 is a flowchart illustrating an example of process priority acquisition processing executed by a VM process priority acquisition unit according to the first embodiment. FIG. 10 is a diagram illustrating a configuration example of a computer system according to a third embodiment. It is a figure which shows an example of the resource usage-amount information of Example 3. 12 is a flowchart illustrating an example of a resource amount calculation process executed by the arrangement management apparatus according to the third embodiment. 12 is a flowchart illustrating an example of a resource amount calculation process executed by the arrangement management apparatus according to the third embodiment. FIG. 10 is a diagram illustrating a configuration example of a computer system according to a fourth embodiment. 14 is a flowchart illustrating an example of process priority acquisition processing executed by an AP process priority acquisition unit according to a fourth embodiment.

  FIG. 1 is a diagram illustrating a configuration example of a computer system according to the first embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of a computer included in the computer system according to the first embodiment.

  The computer system includes an arrangement management device 100, a hypervisor management device 101, a storage device 102, and a plurality of server devices 103. The placement management apparatus 100, the hypervisor management apparatus 101, the storage apparatus 102, and the plurality of server apparatuses 103 are connected to each other via a network.

  The network may be a WAN (Wide Area Network), a LAN (Local Area Network), or the like. Note that the present invention is not limited to the type of network. In addition, the type of network connecting each device may be different. For example, the arrangement management apparatus 100 and the hypervisor management apparatus 101 and a plurality of server apparatuses 103 are connected via a management network, and the plurality of server apparatuses 103 are connected via a business network, and the arrangement management apparatus 100 and the hypervisor are connected. The management apparatus 101, the plurality of server apparatuses 103, and the storage apparatus 102 are connected via a SAN (Storage Area Network).

  A hardware configuration of a computer included in the computer system will be described with reference to FIG. In FIG. 2, the hardware configuration of the computer will be described by taking the arrangement management apparatus 100 as an example.

  The arrangement management apparatus 100 includes a CPU 200, a memory 201, and a NIC 202. The CPU 200, the memory 201, and the NIC 202 are connected to each other via an internal bus or the like.

  CPU 200 executes a program stored in memory 201. The functions of the arrangement management apparatus 100 can be realized by the CPU 200 executing the program. The memory 201 stores a program executed by the CPU 200 and information necessary for executing the program. The memory 201 includes a work area used by the program. The NIC 202 is an interface for connecting to other devices via a network.

  The above is the description of the hardware configuration of the computer included in the computer system. The arrangement management apparatus 100, the hypervisor management apparatus 101, and the server apparatus 103 may have hardware other than the hardware described above. Returning to the description of FIG.

  The server apparatus 103 is a computer on which a VM 170 that executes an OS and an application operates. The memory 201 of the server device 103 stores a program that realizes the hypervisor 150 and stores a data group that is an entity of the VM 170. The present embodiment is not limited to the hypervisor 150. Any program that can realize a general virtualization control unit that manages the VM 170 may be used.

  The hypervisor 150 manages the VM 170. Specifically, the hypervisor 150 generates a VM 170, moves the VM 170, deletes the VM 170, and the like. The hypervisor 150 also manages settings of applications and the like on the VM 170. For example, the hypervisor 150 generates the VM 170 by allocating to the VM 170 a computer resource obtained by logically dividing the computer resource of the server apparatus 103. The hypervisor 150 manages one VM 170 as one process.

  The hypervisor 150 of this embodiment includes a VM process priority acquisition unit 160 as a program module. The VM process priority acquisition unit 160 acquires the process priority of the VM 170 managed as a process by executing a command for acquiring the process priority. Further, when the process priority of the VM 170 is not a numerical value, the VM process priority acquisition unit 160 converts the process priority of the VM 170 into a numerical value.

  The process priority of the VM 170 acquired by the VM process priority acquisition unit 160 is registered in the VM process priority information 132.

  The VM 170 is a virtual computer. The VM 170 includes an OS and an application (not shown). The OS manages one application as one process.

  The VM 170 according to the first embodiment includes an AP process priority acquisition unit 180 as a program module. The AP process priority acquisition unit 180 acquires a process priority of an application managed by the OS by executing a command for acquiring the process priority. If the application process priority is not a numerical value, the AP process priority acquisition unit 180 converts the application process priority into a numerical value.

  The application priority acquired by the AP process priority acquisition unit 180 is registered in the AP process priority information 133.

  The hypervisor management apparatus 101 controls the hypervisor 150 of each server apparatus 103. The memory 201 of the hypervisor management apparatus 101 stores a program that implements the user request reception unit 121 and the hypervisor control unit 122.

  The user request reception unit 121 receives a user request such as an instance movement request. The hypervisor control unit 122 performs various controls on the hypervisor 150.

  The placement management apparatus 100 controls the placement of applications and instances such as the VM 170. The memory 201 of the placement management apparatus 100 stores programs for realizing the placement instruction execution unit 110, the placement destination selection unit 111, and the resource amount calculation unit 112.

  The placement instruction execution unit 110 outputs a command related to instance placement to the hypervisor control unit 122. The placement destination selection unit 111 selects the placement destination of the instance. Specifically, the placement destination selection unit 111 selects the VM 170 in which the application is placed or the hypervisor 150 in which the VM 170 is placed. The resource amount calculation unit 112 calculates the computer resource amount allocated to the application based on the process priority of the process executed on each server device 103.

  The storage device 102 stores various information. The storage apparatus 102 includes a controller, a network interface, and a plurality of storage media (not shown). As the storage medium, HDD (Hard Disk Drive), SSD (Solid State Drive), and the like are considered. The storage apparatus 102 stores server configuration information 131, VM process priority information 132, and AP process priority information 133.

  The server configuration information 131 is information for managing the configuration of the server device 103. Details of the server configuration information 131 will be described with reference to FIG. The VM process priority information 132 is information for managing the process priority of the VM 170. Details of the VM process priority information 132 will be described with reference to FIG. The AP process priority information 133 is information for managing the process priority of the application. Details of the AP process priority information 133 will be described with reference to FIG.

  FIG. 3 is a diagram illustrating an example of the server configuration information 131 according to the first embodiment.

  The server configuration information 131 includes a plurality of entries indicating the configuration of the server device 103. There are as many entries as the number of server apparatuses 103 in the server configuration information 131 of this embodiment. The entries included in the server configuration information 131 include a server device name 301, a hypervisor name 302, a CPU type 303, and a weight 304.

  The server device name 301 is the name of the server device 103 and is identification information for uniquely identifying the server device 103. The hypervisor name 302 is the name of the hypervisor 150, and is identification information for uniquely identifying the hypervisor 150. The CPU type 303 is a type of the CPU 200 that the server apparatus 103 has. A weight 304 is a correction value of the computer resource amount corresponding to the CPU type 303. A process using the weight 304 will be described in a third embodiment.

  FIG. 4 is a diagram illustrating an example of the VM process priority information 132 according to the first embodiment.

  The VM process priority information 132 includes a plurality of entries for managing the process priority of the VM 170 managed by each hypervisor 150. In the VM process priority information 132 of the present embodiment, there are entries for the number of combinations of the hypervisor 150 and the VM 170. The entries included in the VM process priority information 132 include a time stamp 401, a hypervisor name 402, a VM name 403, and a priority 404.

  The time stamp 401 is the time when the process priority of the VM 170 is acquired. The hypervisor name 402 is the same as the hypervisor name 302. The VM name 403 is the name of the VM 170, and is identification information for uniquely identifying the VM 170 managed by the hypervisor 150. The priority 404 is a process priority of the VM 170 corresponding to the VM name 403.

  FIG. 5 is a diagram illustrating an example of the AP process priority information 133 according to the first embodiment.

  The AP process priority information 133 includes a plurality of entries for managing the process priority of the application executed on each VM 170. In the AP process priority information 133 of this embodiment, there are entries for the number of combinations of the hypervisor 150, the VM 170, and the application. The entries included in the AP process priority information 133 include a time stamp 501, a hypervisor name 502, a VM name 503, an AP name 504, and a priority 505.

  The time stamp 501 is the time when the process priority of the application is acquired. The hypervisor name 502 and the VM name 503 are the same as the hypervisor name 302 and the VM name 403. The AP name 504 is the name of an application, and is identification information for uniquely identifying an application executed on the VM 170. The priority 505 is the process priority of the application corresponding to the AP name 504.

  FIG. 6 is a sequence diagram illustrating the flow of the instance arrangement process in the computer system according to the first embodiment. FIG. 7 is a diagram illustrating an example of the input screen according to the first embodiment.

  The hypervisor management apparatus 101 receives an input from the user (step S101).

  Specifically, the user request receiving unit 121 displays an input screen 700 as shown in FIG. 7 and receives an input from the user. The user request receiving unit 121 outputs the input information including the value input on the input screen 700. The input information according to the first embodiment includes the name of the instance whose arrangement is to be changed, the SLA of the application that is the target of the SLA, and an instruction to automate the movement of the instance.

  Here, the input screen 700 will be described. The input screen 700 is displayed by the user request receiving unit 121. The input screen 700 includes input fields 701, 703, and 704, a check field 702, an OK button 705, and a cancel button 706.

  The input column 701 is a column for inputting the name of the instance whose arrangement is to be changed. In this embodiment, the name of the VM 170 or the name of the application is entered in the input field 701. When the name of the application is input in the input field 701, it means that the VM 170 that executes the application is changed.

  The check column 702 is a column that permits automation of the movement of instances between the hypervisors 150. In the first embodiment, it is assumed that the check field 702 is operated.

  The input field 703 is a field for inputting the name of the application targeted for SLA. An input field 703 displays values that can be selected from a pull-down list. When an application name is input in the input field 701, the input field 703 may be blank because the application is an application that is a target of SLA.

  The input column 704 is a column for inputting an SLA to be satisfied. In this embodiment, the computer resource amount allocated to the application from the server apparatus 103 is input as an SLA to be satisfied. The input column 704 displays values that can be selected from the pull-down list.

  An OK button 705 is an operation button for confirming the value input on the input screen 700 and registering it in the hypervisor management apparatus 101. A cancel button 706 is an operation button for canceling confirmation of the value input on the input screen 700.

  The above is the description of the input screen 700. Returning to the description of FIG.

  The hypervisor management apparatus 101 transmits an instance movement request including input information to the arrangement management apparatus 100 (step S102).

  When the placement management apparatus 100 receives an instance move request, the placement management apparatus 100 executes placement destination determination processing (step S103). Details of the placement destination determination process will be described with reference to FIG.

  The placement management apparatus 100 transmits a response including the processing result of the placement destination determination process to the hypervisor management apparatus 101 (step S104).

  When receiving a response, the hypervisor management apparatus 101 executes an arrangement process (step S105).

  Specifically, when the information on the instance to be moved is included in the response, the hypervisor control unit 122 issues an instance placement instruction including the information on the instance. When information indicating an error is included in the response, the hypervisor control unit 122 notifies the user of the error.

  FIG. 8 is a flowchart illustrating an example of an arrangement destination determination process executed by the arrangement management apparatus 100 according to the first embodiment.

  When the placement management apparatus 100 receives an instance move request, the placement management apparatus 100 calls the placement command execution unit 110 to start placement destination determination processing.

  The placement instruction execution unit 110 calls the placement destination selection unit 111 and instructs execution of the placement destination selection process (step S201). Details of the placement destination selection process will be described with reference to FIG.

  The placement instruction execution unit 110 waits until a processing result is input from the placement destination selection unit 111. When the processing result is input from the placement destination selection unit 111, the placement instruction execution unit 110 determines whether or not the hypervisor 150 that is the placement destination of the instance has been selected (step S202).

  Specifically, the placement instruction execution unit 110 determines whether or not the identification information (name) of the hypervisor 150 is included in the processing result. When the identification information of the hypervisor 150 is included in the processing result, the placement instruction execution unit 110 determines that the hypervisor 150 that is the placement destination of the instance has been selected.

  If it is determined that the hypervisor 150 that is the instance placement destination is selected, the placement instruction execution unit 110 transmits a response including the processing result of the placement destination determination process to the hypervisor management apparatus 101 (step S203).

  The processing result of the placement destination determination process includes the identification information of the placement destination hypervisor 150, the identification information of the instance whose placement is to be changed, and the like. For example, when the instance whose placement is to be changed is the VM 170, the processing result includes the name of the placement destination hypervisor 150 and the name of the VM 170. When the instance whose arrangement is to be changed is an application, the processing result includes the name of the VM at the arrangement destination, the name of the hypervisor 150 that manages the VM, and the name of the application.

  If it is determined that the hypervisor 150 that is the instance placement destination is not selected, the placement instruction execution unit 110 transmits a response including information indicating an error to the hypervisor management apparatus 101 (step S204).

  FIG. 9 is a flowchart illustrating an example of an arrangement destination selection process executed by the arrangement management apparatus 100 according to the first embodiment.

  When the placement destination selection unit 111 is instructed to execute the placement destination selection process by the placement instruction execution unit 110, the placement destination selection unit 111 starts the placement destination selection process. At this time, the placement instruction execution unit 110 inputs input information included in the instance movement request to the placement destination selection unit 111.

  The placement destination selection unit 111 calls the resource amount calculation unit 112 and instructs the execution of the resource amount calculation process (step S301). The placement destination selection unit 111 waits until a processing result is input from the resource amount calculation unit 112. Details of the resource amount calculation processing will be described with reference to FIG.

  The processing result includes a list of entries including the name of the hypervisor 150 and the amount of computer resources allocated from the server apparatus 103 in which the hypervisor 150 operates to the target application of the SLA. In the following description, the computer resource amount allocated to the SLA target application from the server apparatus 103 on which the hypervisor 150 operates is also referred to as an allocated resource amount. The allocated resource amount represents a predicted value of the computer resource amount allocated to the target application of the SLA.

  In this embodiment, the allocation rate of the computer resource amount is calculated as the allocated resource amount. As will be described later, this embodiment is characterized in that the allocated resource amount is calculated based on the priority of the application and the priority of the VM 170.

  When the processing result is input from the resource amount calculation unit 112, the placement destination selection unit 111 determines whether there is a hypervisor 150 that satisfies the SLA based on the SLA value and the processing result (step S302).

  Specifically, the placement destination selection unit 111 determines whether or not an entry whose allocated resource amount is larger than the SLA value is included in the list. For example, when the SLA value illustrated in FIG. 8 is input, the placement destination selection unit 111 determines whether or not the list includes an entry having a CPU resource amount allocation rate of “0.05” or more.

  When it is determined that there is a hypervisor 150 that satisfies the SAL, the placement destination selection unit 111 selects one hypervisor 150 that places an instance (step S303).

  For example, the placement destination selection unit 111 selects the hypervisor 150 having the largest allocation rate of the computer resource amount from the hypervisors 150 satisfying the SLA.

  The placement destination selection unit 111 outputs a processing result including the name of the selected hypervisor 150 to the placement instruction execution unit 110 (step S304).

  If it is determined in step S302 that there is no hypervisor 150 that satisfies the SLA, the placement destination selection unit 111 outputs a processing result including an error to the placement instruction execution unit 110 (step S305).

  10A and 10B are flowcharts illustrating an example of a resource amount calculation process executed by the arrangement management apparatus 100 according to the first embodiment.

  When the resource amount calculation unit 112 is instructed to execute the resource amount calculation process from the placement destination selection unit 111, the resource amount calculation unit 112 starts the resource amount calculation process. At this time, the placement destination selection unit 111 inputs the name of the instance whose placement is to be changed to the resource amount calculation unit 112.

  The resource amount calculation unit 112 determines whether the instance whose arrangement is to be changed is the VM 170 (step S401).

  Specifically, it is determined whether or not the name of the instance whose arrangement is to be changed is the name of the VM 170. When the name of the instance whose arrangement is to be changed is the name of the application, the resource amount calculation unit 112 determines that the instance whose arrangement is to be changed is not the VM 170. In the following description, the VM 170 whose arrangement is changed is also referred to as a target VM 170. In the following description, an application whose arrangement is to be changed is also referred to as a target application.

  When it is determined that the instance whose arrangement is to be changed is the VM 170, the resource amount calculation unit 112 acquires the priorities of all the VMs 170 from the VM process priority information 132, and all of the VM process priority information 133 from the AP process priority information 133 The priority of the application is acquired (step S402).

  Specifically, the resource amount calculation unit 112 refers to the VM process priority information 132 and acquires the latest priority of each VM 170. Further, the resource amount calculation unit 112 refers to the AP process priority information 133 and acquires the latest priority of each application.

  The resource amount calculation unit 112 refers to the VM process priority information 132 to obtain the maximum priority value of each VM 170 and refers to the AP process priority information 133 to determine the maximum priority of each application. A value may be acquired.

  Next, the resource amount calculation unit 112 selects the target hypervisor 150 from the plurality of hypervisors 150 (step S403). In addition, the process of step S402 may be performed after the process of step S403 is performed. Thereby, the amount of data to be acquired can be reduced.

  Next, when the target VM 170 is moved to the target hypervisor 150, the resource amount calculation unit 112 calculates a computer resource amount allocated from the target hypervisor 150 to the target VM 170 (step S404).

  Specifically, the resource amount calculation unit 112 substitutes the process priority of the target VM 170 and the process priorities of all the VMs 170 managed by the target hypervisor 150 into the following formula (1). In the following description, the value calculated in step S404 is also described as the first resource amount.

  Next, based on the information acquired from the AP process priority information 133, the resource amount calculation unit 112 calculates the computer resource amount allocated to the target application of the SLA from the target VM 170 (step S405).

  Specifically, the resource amount calculation unit 112 substitutes the process priority of the application targeted for SLA and the process priority of all applications on the target VM 170 including the application targeted for SLA into the following formula (2). . In the following description, the value calculated in step S405 is also described as the second resource amount.

  Next, the resource amount calculation unit 112 multiplies the first resource amount and the second resource amount to multiply the computer resource amount (allocated resource) allocated from the server device 103 in which the target hypervisor 150 is operated to the target application of the SLA. Amount) is calculated (step S406).

  In step S 406, the resource amount calculation unit 112 stores a list of entries including the name of the target hypervisor 150 and the allocated resource amount in the memory 201.

  Next, the resource amount calculation unit 112 determines whether or not the processing has been completed for all the hypervisors 150 (step S407).

  When it is determined that the processing has not been completed for all the hypervisors 150, the resource amount calculation unit 112 returns to step S403 and executes the same processing.

  When it is determined that the processing has been completed for all the hypervisors 150, the resource amount calculation unit 112 outputs the list to the placement destination selection unit 111 as a processing result (step S408).

  When it is determined in step S401 that the instance whose arrangement is to be changed is an application, the resource amount calculation unit 112 acquires the priorities of all the applications from the AP process priority information 133 (step S411).

  Next, the resource amount calculation unit 112 selects the target hypervisor 150 from the plurality of hypervisors 150 (step S412), and selects the target VM 170 from the VMs 170 managed by the target hypervisor 150 (step S413). . In addition, the process of step S411 may be performed after the process of step S413 is performed. Thereby, the amount of data to be acquired can be reduced.

  Next, the resource amount calculation unit 112 calculates the resource amount allocated to the target application from the target VM 170 (step S414).

  Specifically, the resource amount calculation unit 112 substitutes the process priority of the application targeted for SLA and the process priority of the application executed on the target VM 170 into the following formula (3). The process priority of the SLA target application is the process priority on the VM 170 that executes the SLA target application. In the following description, the value calculated in step S414 is also described as the third resource amount.

  Next, the resource amount calculation unit 112 calculates the third resource amount as the allocated resource amount (step S415). In step S 415, the resource amount calculation unit 112 stores a list of entries including the name of the target hypervisor 150 and the allocated resource amount in the memory 201. Next, the resource amount calculation unit 112 determines whether the processing has been completed for all VMs 170 managed by the target hypervisor 150 (step S416).

  When it is determined that the processing has not been completed for all the VMs 170 managed by the target hypervisor 150, the resource amount calculation unit 112 returns to step S413 and executes the same processing.

  When it is determined that processing has been completed for all VMs 170 managed by the target hypervisor 150, the resource amount calculation unit 112 determines whether processing has been completed for all hypervisors 150 (step S416). The process in step S416 is the same as the process in step S407.

  If it is determined that the processing has not been completed for all the hypervisors 150, the resource amount calculation unit 112 returns to step S412 and executes the same processing.

  When it is determined that the processing has been completed for all the hypervisors 150, the resource amount calculation unit 112 outputs the list to the placement destination selection unit 111 as a processing result (step S408).

  As described with reference to FIGS. 10A and 10B, in the resource amount calculation process, the allocated resource amount is calculated based on the process priority of the application and the process priority of the VM 170. Thereby, the placement management apparatus 100 can search for the hypervisor 150 that is the placement destination of the instance based on the allocated resource amount.

  FIG. 11 is a flowchart illustrating an example of process priority acquisition processing executed by the VM process priority acquisition unit 160 according to the first embodiment.

  The VM process priority acquisition unit 160 executes process priority acquisition processing periodically or when an instance migration request is received.

  The VM process priority acquisition unit 160 executes a command for acquiring the process priority (step S501). Since the command for obtaining the process priority is a known command, the description thereof is omitted.

  When the acquired process priority is a value other than a numerical value, the VM process priority acquisition unit 160 converts the acquired value into a numerical value.

  Next, the VM process priority acquisition unit 160 calculates a relative value of the acquired process priority (step S502).

  Specifically, the VM process priority acquisition unit 160 calculates the relative value of the process priority of each VM 170 using the following formula (4) or the following formula (5).

  Next, the VM process priority acquisition unit 160 generates registration information (step S503).

  Specifically, the VM process priority acquisition unit 160 generates one entry including a time stamp, the name of the hypervisor 150, the name of the VM 170, and the relative value of the process priority of the VM 170 for each VM 170. . Further, the VM process priority acquisition unit 160 generates registration information including a plurality of entries and a registration instruction.

  Next, the VM process priority acquisition unit 160 transmits information for registration to the storage apparatus 102 (step S504). Thereafter, the VM process priority acquisition unit 160 waits for a predetermined time (step S505).

  Next, the VM process priority acquisition unit 160 determines whether an end command has been received (step S506).

  When it is determined that the end command has not been received, the VM process priority acquisition unit 160 returns to Step S501 and executes the same processing.

  When it is determined that the end command has been received, the VM process priority acquisition unit 160 ends the process priority acquisition processing.

  The process priority acquisition process executed by the AP process priority acquisition unit 180 is the same as the process priority acquisition process shown in FIG. However, the processing contents are different. Specific processing contents will be described below.

  In step S501, the AP process priority acquisition unit 180 executes a command for acquiring a process priority. When the acquired process priority is a value other than a numerical value, the AP process priority acquisition unit 180 converts the acquired value into a numerical value.

  In step S502, the AP process priority acquisition unit 180 calculates the relative value of the process priority of each application using the following expression (6) or the following expression (7).

  In step S503, the AP process priority acquisition unit 180 generates one entry including a time stamp, the name of the hypervisor 150, the name of the VM 170, the name of the application, and the relative value of the process priority of the application. Further, the AP process priority acquisition unit 180 generates registration information including a plurality of entries and a registration instruction.

  In step S <b> 504, the AP process priority acquisition unit 180 transmits registration information to the storage apparatus 102.

  Since step S505 and step S506 are the same processing, description is abbreviate | omitted.

  In the first embodiment, the allocated resource amount is calculated based on the process priority. However, the present invention is not limited to this. For example, a method of calculating the allocated resource amount based on the length of the process queue or the like is also conceivable.

  According to the first embodiment, when the instance on the server apparatus 103 is migrated to another server apparatus 103, the arrangement management apparatus 100 considers the amount of computer resources actually allocated to the application, that is, the SLA, and the arrangement destination Hypervisor 150 (server apparatus 103) is selected. As a result, the SLA of the application can be satisfied even at the placement destination.

  The second embodiment is different from the first embodiment in that a process priority statistical value is used. Hereinafter, the second embodiment will be described focusing on differences from the first embodiment.

  The configuration of the computer system of the second embodiment is the same as that of the first embodiment. The hardware configuration and software configuration of each device of the second embodiment are the same as those of the first embodiment. The various information of the second embodiment is the same as that of the first embodiment. The flow of the instance arrangement process of the second embodiment is the same as that of the first embodiment. The placement destination determination process and the placement destination selection process in the second embodiment are the same as those in the first embodiment.

  The resource amount calculation process according to the second embodiment is partially different from the resource amount calculation process according to the first embodiment.

  The process in step S401 is the same process as in the first embodiment.

  In step S402, the resource amount calculation unit 112 acquires a process priority history of each VM 170. That is, the resource amount calculation unit 112 acquires a plurality of entries with different time stamps for one VM 170. The resource amount calculation unit 112 calculates the average value of the process priority of each VM 170 using the history of each VM 170.

  In step S402, the resource amount calculation unit 112 acquires a process priority history of each application. That is, the resource amount calculation unit 112 acquires a plurality of entries having different time stamps for one application. The resource amount calculation unit 112 calculates an average value of process priorities of each application.

  The process in step S403 is the same as that in the first embodiment.

  In step S404, the resource amount calculation unit 112 calculates the first resource amount using the average value of the process priority of the VM 170. In step S405, the resource amount calculation unit 112 calculates the second resource amount using the average value of the process priority of the application.

  The processing from step S405 to step S408 is the same processing as in the first embodiment.

  In step S411, the resource amount calculation unit 112 acquires a process priority history of each application. The resource amount calculation unit 112 calculates an average value of process priorities of each application.

  The processes in steps S412 and S413 are the same as those in the first embodiment.

  In step S414, the resource amount calculation unit 112 calculates the third resource amount using the average value of the process priority of the application.

  The processing from step S415 to step S417 is the same processing as in the first embodiment.

  According to the second embodiment, by using the history, it is possible to select the hypervisor 150 that is the placement destination of the instance that satisfies the SLA of the application more accurately.

  The third embodiment is different from the first embodiment in that the use amount of the computer resource of the server apparatus 103 is also considered. Hereinafter, the third embodiment will be described focusing on differences from the first embodiment.

  FIG. 12 is a diagram illustrating a configuration example of a computer system according to the third embodiment.

  The hardware configuration of each device of the third embodiment is the same as that of the first embodiment. The software configurations of the arrangement management apparatus 100 and the hypervisor management apparatus 101 of the third embodiment are the same as those of the first embodiment.

  The server apparatus 103 according to the third embodiment is different in software configuration from the server apparatus 103 according to the first embodiment. Specifically, the hypervisor 150 includes a computer resource usage measuring unit 1200 as a module. In addition, the storage apparatus 102 according to the third embodiment includes resource usage information 1211.

  The computer resource usage measurement unit 1200 measures the usage of the computer resource that the server apparatus 103 has. The computer resource usage measured by the computer resource usage measuring unit 1200 is registered in the resource usage information 1211.

  The resource usage information 1211 is information for managing the usage of the computer resources that the server apparatus 103 has. FIG. 13 is a diagram illustrating an example of the resource usage information 1211 according to the third embodiment.

  The resource usage information 1211 includes a plurality of entries for managing the usage rate of the CPU 200 included in each server device 103. Each entry includes a time stamp 1301, a server device name 1302, and a CPU usage rate 1303.

  The time stamp 1301 is the time when the CPU usage rate is acquired. The server device name 1302 is the same as the server device name 301. The CPU usage rate 1303 is a usage rate of the CPU 200 included in the server device 103.

  The resource usage information 1211 may include the usage amount of the memory 201, the I / O usage amount, the bandwidth usage amount, and the like.

  The flow of the instance arrangement process of the third embodiment is the same as that of the first embodiment. The placement destination determination process and the placement destination selection process in the third embodiment are the same as those in the first embodiment.

  The resource amount calculation process according to the third embodiment is partially different from the resource amount calculation process according to the first embodiment.

  14A and 14B are flowcharts illustrating an example of a resource amount calculation process executed by the arrangement management apparatus 100 according to the third embodiment.

  The processes in steps S401 and S402 are the same as those in the first embodiment.

  After the process of step S402 is executed, the resource amount calculation unit 112 acquires the weights of all the server devices 103 from the server configuration information 131, and also calculates the computer resource information of all the server devices 103 from the resource usage information 1211. The usage amount is acquired (step S451).

  The processing from step S403 to step S405 is the same as the processing in the first embodiment.

  After the processing of step S405 is executed, the resource amount calculation unit 112 determines the first resource amount, the second resource amount, the usage amount of the computer resource of the server device 103 on which the target hypervisor 150 operates, and the server device 103 The allocation resource amount is calculated by multiplying the weight 304 (step S452). That is, the allocated resource amount is corrected based on the usage amount of the computer resource of the server device 103 and the weight 304 of the server device 103.

  The processes in steps S407 and S408 are the same as those in the first embodiment. Further, the processing in step S411 is the same as the processing in the first embodiment.

  After the processing of step S411 is executed, the resource amount calculation unit 112 acquires the weights of all the server devices 103 from the server configuration information 131, and the computer resource information of all the server devices 103 from the resource usage information 1211. The usage amount is acquired (step S453). The process of step S453 is the same as the process of step S451.

  The processing from step S412 to step S414 is the same as the processing in the first embodiment.

  After the process of step S414 is executed, the resource amount calculation unit 112 multiplies the third resource amount, the usage amount of the computer resource of the server device 103 on which the target hypervisor 150 operates, and the weight 304 of the server device 103. Thus, the allocated resource amount is calculated (step S454). That is, the allocated resource amount is corrected based on the usage amount of the computer resource of the server device 103 and the weight 304 of the server device 103.

  The processes in steps S416 and S417 are the same as those in the first embodiment.

  According to the third embodiment, the placement management apparatus 100 selects the placement destination server apparatus 103 in consideration of not only the priority of the application and the VM 170 but also the performance and usage of the computer resource of the server apparatus 103. Can do. As a result, it is possible to select the hypervisor 150 that is the placement destination of the instance that satisfies the SLA of the application more accurately.

  The fourth embodiment is different from the first embodiment in that the hypervisor 150 acquires the process priority of the application. Hereinafter, the fourth embodiment will be described focusing on differences from the first embodiment.

  FIG. 15 is a diagram illustrating a configuration example of a computer system according to the fourth embodiment.

  The hardware configuration of each device of the fourth embodiment is the same as that of the first embodiment. The software configurations of the arrangement management device 100 and the hypervisor management device 101 of the fourth embodiment are the same as those of the first embodiment. Various information of the fourth embodiment is the same as that of the first embodiment.

  The server device 103 according to the fourth embodiment is different in software configuration from the server device 103 according to the first embodiment. Specifically, the hypervisor 150 includes an AP process priority acquisition unit 1500 as a module.

  The flow of the instance arrangement process of the fourth embodiment is the same as that of the first embodiment. The placement destination determination process, placement destination selection process, and resource amount calculation process of the fourth embodiment are the same as those of the first embodiment.

  In the fourth embodiment, the process priority acquisition process executed by the AP process priority acquisition unit 1500 is partially different. FIG. 16 is a flowchart illustrating an example of process priority acquisition processing executed by the AP process priority acquisition unit 1500 according to the fourth embodiment.

  The AP process priority acquisition unit 1500 acquires the process priority of the application by referring to the storage area of the memory 201 allocated to each VM 170 instead of executing the command (step S551).

  The processing from step S502 to step S506 is the same as the processing in the first embodiment.

  In the fourth embodiment, the processing load can be reduced because the VM 170 does not acquire the process priority of the application. Further, the fourth embodiment has the same effect as the first embodiment.

  In addition, the numerical formula shown in this specification is an example, and is not limited to this.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. Further, for example, the above-described embodiments are described in detail for easy understanding of the present invention, and are not necessarily limited to those provided with all the described configurations. In addition, a part of the configuration of the embodiment can be added to, deleted from, or replaced with another configuration.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. The present invention can also be realized by software program codes that implement the functions of the embodiments. In this case, a storage medium in which the program code is recorded is provided to the computer, and a CPU included in the computer reads the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code itself and the storage medium storing it constitute the present invention. As a storage medium for supplying such a program code, for example, a flexible disk, a CD-ROM, a DVD-ROM, a hard disk, an SSD (Solid State Drive), an optical disk, a magneto-optical disk, a CD-R, a magnetic tape, A non-volatile memory card, ROM, or the like is used.

  Further, the program code for realizing the functions described in the present embodiment can be implemented by a wide range of programs or script languages such as assembler, C / C ++, perl, Shell, PHP, Java, and the like.

  Furthermore, by distributing the program code of the software that implements the functions of the embodiments via a network, the program code is stored in a storage means such as a hard disk or memory of a computer or a storage medium such as a CD-RW or CD-R. The CPU included in the computer may read and execute the program code stored in the storage unit or the storage medium.

  In the above-described embodiments, the control lines and information lines indicate what is considered necessary for the explanation, and not all control lines and information lines on the product are necessarily shown. All the components may be connected to each other.

Claims (15)

A computer system comprising a plurality of computers,
The plurality of computers have a processor, a memory connected to the processor, an interface connected to the processor,
The plurality of computers includes a plurality of business computers in which a virtualization control unit that manages virtual computers operates, a placement management computer that manages the placement of the virtual computers that execute applications, and virtualization that controls the virtualization control unit Including control unit management computer,
The arrangement management computer is
Assigned to the service quality assurance target application from the virtualization control unit in which the virtual machine executing the service quality assurance target application is executed based on the process priority of the process executed on the business computer A resource amount calculator that calculates an allocated resource amount that is a predicted value of the computer resource amount
Based on the allocated resource amount, the virtualization control unit that is the placement destination of a virtual machine that executes the application for which the service quality is to be guaranteed is selected, and the service quality assurance target is selected for the selected virtualization control unit. A placement instruction execution unit that instructs placement of a virtual machine that executes the application of
The virtualization control unit management computer is configured to instruct the selected virtualization control unit to set a virtual computer that executes the service quality assurance target application based on an instruction from the placement instruction execution unit. A computer system characterized by comprising: The computer system according to claim 1,
The placement management computer manages first process priority information for managing process priority of the application on the virtual computer, and process priority of the virtual computer managed by the virtualization control unit. Manage second process priority information,
The resource amount calculation unit
When an instance move request including identification information of a migration target virtual machine, identification information of the service quality assurance target application executed on the migration target virtual machine, and a value of the service quality assurance is received, Select a target virtualization control unit from the plurality of virtualization control units,
With reference to the second process priority information, the process priority of the migration target virtual machine and the process priority of the virtual machine managed by the target virtualization control unit are acquired,
With reference to the first process priority information, obtain the process priority of the application executed by the virtual machine to be moved,
Based on the process priority of the migration target virtual machine and the process priority of the virtual machine managed by the target virtualization control unit, the computer assigned to the migration target virtual machine from the target virtualization control unit Calculate the first resource amount, which is the resource amount,
Based on the process priority of the application executed by the migration target virtual computer, a second resource amount that is a computer resource amount allocated from the migration target virtual computer to the service quality assurance target application is calculated;
Calculating the allocated resource amount based on the first resource amount and the second resource amount;
The computer system characterized in that the placement instruction execution unit selects the virtualization control unit based on the allocated resource amount and the service quality assurance value. The computer system according to claim 1,
The placement management computer manages first process priority information for managing process priority of the application on the virtual computer;
The resource amount calculation unit
When an instance move request that includes the identification information of the service quality assurance target application and the service quality assurance value is received, the virtual object of the target is selected from the plurality of virtualization control units. Select the control unit,
Select a target virtual machine from at least one virtual machine managed by the target virtualization control unit,
With reference to the first process priority information, obtain the process priority of the service quality assurance target application and the process priority of the application executed by the target virtual machine,
Based on the process priority of the target application of the service quality guarantee and the process priority of the application executed by the target virtual machine, the allocated resource amount is calculated,
The computer system characterized in that the placement instruction execution unit selects the virtualization control unit based on the allocated resource amount and the service quality assurance value. A computer system according to claim 2 or claim 3, wherein
The placement instruction execution unit selects the virtualization control unit in which the allocated resource amount is equal to or greater than the service quality assurance value and the computer resource amount allocated to the service quality assurance target application is the largest. A computer system characterized by that. A computer system according to claim 4, wherein
The virtualization control unit includes a first process priority acquisition unit that acquires a process priority of the virtual machine,
At least one of the virtualization control unit and the virtual computer includes a second process priority acquisition unit that acquires a process priority of the application. A computer system according to claim 4, wherein
The placement management computer manages configuration information for managing performance values of computer resources possessed by the plurality of business computers, and resource usage information for managing usage of the computer resources,
The resource amount calculation unit
Referring to the configuration information, obtain the performance value of the computer resource of the business computer on which the target virtualization control unit operates,
Referring to the resource usage information, obtain the usage amount of the computer resource in the business computer on which the target virtualization control unit operates,
The allocation based on the performance value of the computer resource of the business computer on which the target virtualization control unit operates and the usage amount of the computer resource amount on the business computer on which the target virtualization control unit operates Correct the resource amount,
The computer system characterized in that the placement instruction execution unit selects the virtualization control unit based on the corrected allocated resource amount. A computer system control method comprising a plurality of computers,
The plurality of computers have a processor, a memory connected to the processor, an interface connected to the processor,
The plurality of computers includes a plurality of business computers in which a virtualization control unit that manages virtual computers operates, a placement management computer that manages the placement of the virtual computers that execute applications, and virtualization that controls the virtualization control unit Including control unit management computer,
The computer system control method includes:
Based on the process priority of the process executed on the business computer, the placement management computer receives the service quality assurance from the virtualization control unit in which the virtual computer that executes the service quality assurance target application is placed. A first step of calculating an allocated resource amount that is a predicted value of a computer resource amount allocated to the target application;
A second step in which the placement management computer selects, based on the allocated resource amount, the virtualization control unit to be a placement destination of a virtual computer that executes the application of the service quality assurance target;
A third step in which the placement management computer instructs the selected virtualization control unit to place a virtual computer that executes the service quality assurance target application;
The virtualization control unit management computer instructs the selected virtualization control unit to set a virtual computer that executes the service quality assurance target application based on an instruction from the placement management computer. And a step of controlling the computer system. A control method for a computer system according to claim 7,
The placement management computer manages first process priority information for managing process priority of the application on the virtual computer, and process priority of the virtual computer managed by the virtualization control unit. Manage second process priority information,
The first step includes
Movement of an instance in which the placement management computer includes identification information of a migration target virtual machine, identification information of the service quality assurance target application executed on the migration target virtual machine, and a value of the service quality assurance When receiving the request, selecting a target virtualization control unit from among the plurality of virtualization control units;
The placement management computer refers to the second process priority information, and acquires the process priority of the migration target virtual computer and the process priority of the virtual computer managed by the target virtualization control unit When,
The placement management computer refers to the first process priority information to obtain the process priority of the application executed by the migration target virtual machine;
Based on the process priority of the migration target virtual machine and the process priority of the virtual machine managed by the target virtualization control unit, the placement management computer from the target virtualization control unit Calculating a first resource amount that is a computer resource amount allocated to the virtual machine;
A second resource that is a computer resource amount that the placement management computer allocates from the migration target virtual computer to the service quality assurance target application based on the process priority of the application executed by the migration target virtual computer Calculating a quantity;
The placement management computer calculating the allocated resource amount based on the first resource amount and the second resource amount;
In the second step, the arrangement management computer selects the virtualization control unit based on the allocated resource amount and the service quality assurance value. A control method for a computer system according to claim 7,
The placement management computer manages first process priority information for managing process priority of the application on the virtual computer;
The first step includes
When the placement management computer receives an instance movement request that includes the identification information of the application that is the movement target and the service quality assurance target and the service quality assurance value, a plurality of the virtualization control units Selecting a target virtualization control unit from
The placement management computer selecting a target virtual computer from at least one of the virtual computers managed by the target virtualization control unit;
The placement management computer refers to the first process priority information, and acquires the process priority of the service quality assurance target application and the process priority of the application executed by the target virtual computer. When,
The placement management computer calculating the allocated resource amount based on the process priority of the application targeted for the service quality assurance and the process priority of the application executed by the target virtual computer;
In the second step, the arrangement management computer selects the virtualization control unit based on the allocated resource amount and the service quality assurance value. A control method for a computer system according to claim 8 or 9, wherein
In the second step, the placement management computer has the virtual resource in which the allocated resource amount is equal to or greater than the service quality assurance value and the computer resource amount allocated to the service quality assurance target application is the largest. A computer system control method comprising: selecting a computer control unit. A control method for a computer system according to claim 10,
The computer system control method includes:
The virtualization control unit obtaining a process priority of the virtual machine;
A method for controlling a computer system, comprising: at least one of the virtualization control unit and the virtual computer acquiring a process priority of the application. A control method for a computer system according to claim 10,
The placement management computer manages configuration information for managing performance values of computer resources possessed by the plurality of business computers, and resource usage information for managing usage of the computer resources,
The first step includes
The placement management computer refers to the configuration information, and obtains a performance value of the computer resource of the business computer on which the target virtualization control unit operates;
The placement management computer refers to the resource usage information and obtains the usage amount of the computer resource in the business computer on which the target virtualization control unit operates;
The placement management computer has a performance value of the computer resource of the business computer in which the target virtualization control unit operates, and a usage amount of the computer resource amount in the business computer in which the target virtualization control unit operates. Correcting the allocated resource amount based on:
In the second step, the placement management computer selects the virtualization control unit based on the corrected allocated resource amount. A program executed by a placement management computer connected to a plurality of computers on which a virtualization control unit that manages virtual computers operates,
The arrangement management computer has a processor, a memory connected to the processor, and an interface connected to the processor.
The program is
Assigned to the service quality assurance target application from the virtualization control unit in which the virtual machine executing the service quality assurance target application is executed based on the process priority of the process executed on the business computer A first procedure for calculating an allocated resource amount that is a predicted value of the computer resource amount to be executed;
A second procedure of selecting the virtualization control unit to be a placement destination of a virtual machine that executes the application of the service quality assurance target based on the allocated resource amount;
A third procedure for instructing the selected virtualization control unit to place a virtual machine that executes the application of the service quality assurance target;
Instructing the selected virtualization control unit to set a virtual machine that executes the service quality assurance target application based on an instruction to place a virtual machine that executes the service quality assurance target application And causing the arrangement management computer to execute the above procedure. The program according to claim 13,
The placement management computer manages first process priority information for managing process priority of the application on the virtual computer, and process priority of the virtual computer managed by the virtualization control unit. Manage second process priority information,
The first procedure includes:
When an instance move request including identification information of a migration target virtual machine, identification information of the service quality assurance target application executed on the migration target virtual machine, and a value of the service quality assurance is received, A procedure for selecting a target virtualization control unit from the plurality of virtualization control units;
A step of referring to the second process priority information to obtain a process priority of the migration target virtual machine and a process priority of the virtual machine managed by the target virtualization control unit;
A procedure for referring to the first process priority information to obtain a process priority of the application executed by the virtual machine to be migrated;
Based on the process priority of the migration target virtual machine and the process priority of the virtual machine managed by the target virtualization control unit, the computer assigned to the migration target virtual machine from the target virtualization control unit A procedure for calculating a first resource amount which is a resource amount;
Calculating a second resource amount that is a computer resource amount allocated to the service quality assurance target application from the migration target virtual computer based on the process priority of the application executed by the migration target virtual computer; ,
Calculating the allocated resource amount based on the first resource amount and the second resource amount;
The second procedure includes a procedure for selecting the virtualization control unit based on the allocated resource amount and the service quality assurance value. The program according to claim 13,
The placement management computer manages first process priority information for managing process priority of the application on the virtual computer;
The first procedure includes:
When an instance move request that includes the identification information of the service quality assurance target application and the service quality assurance value is received, the virtual object of the target is selected from the plurality of virtualization control units. A procedure for selecting a control unit,
A procedure for selecting a target virtual machine from at least one of the virtual machines managed by the target virtualization control unit;
A step of referring to the first process priority information to obtain a process priority of the target application of the service quality guarantee and a process priority of the application executed by the target virtual machine;
Calculating the allocated resource amount based on the process priority of the application subject to the service quality assurance and the process priority of the application executed by the target virtual machine,
The second procedure includes a procedure for selecting the virtualization control unit based on the allocated resource amount and the service quality assurance value.

Images