JP2010113677A - Service management device, service management method, and service management system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To determine a migration destination of a virtual machine on the basis of the resource usage of a service in a virtual system. <P>SOLUTION: A service management device 2 determines a migration destination of the virtual machine and migrates it so that the service operating on the virtual machine satisfies a condition that the service is completed by a completion deadline on the basis of service configuration information, job management information, host machine configuration information, virtual machine configuration information, resource management information, and event information. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for moving a virtual machine in a virtual system environment.

  In recent years, virtual systems have attracted attention from the viewpoints of IT (Information Technology) operation cost reduction, ease of operation management, and service continuity. In the virtualization system, the service is provided without being aware of the configuration of the business computer on which the service is operating. Furthermore, by allocating a resource that can be shared in the business computer in an arbitrary size for each virtual machine and using it as a virtual resource, a plurality of virtual machines having performance differences can be constructed in the same computer. Even after the virtual machine is constructed, the shared resource allocation can be dynamically changed, and the services in the virtual machine can be provided independently.

  In addition, when a service cannot be continued with a virtual machine, such as when a failure occurs in a business computer resource that runs the virtual machine, a business program that configures the service in the virtual machine is deployed on another business computer. By moving the virtual machine in this state, services by the virtual machine can be continuously provided.

Further, according to the technique disclosed in Patent Document 1, it is possible to dynamically allocate resources in the business computer according to the performance target of the service.
In addition, with regard to the amount of resources used by services in business computers and virtual machines, OS (Operating System) and resource monitoring products are provided by various companies. Resource usage can be obtained for each virtual machine.
JP 2007-48315 A

  In a virtualized system, the virtual machine that provides the service can be moved to another business computer, and the amount of resources allocated to the virtual machine can be changed dynamically. It is extremely difficult to specify the business computer resource actually allocated from the resource amount. In addition, because the shared resources of business computers are allocated to any virtual machine, the virtual resources built on business computers can be used to obtain the shared resource amounts actually used by services on the virtual machines from the usage of shared resources. Information about services running on machines and virtual machines is required.

  Also, when a shared resource in a business computer fails and the virtual machine to which the failed shared resource is assigned needs to be moved to another business computer, it was running on the target virtual machine It is necessary to move the virtual machine to another business computer that can allocate the amount of resources used by the service to the virtual machine. Here, if the service to be moved is composed of multiple jobs such as one-day order clearing work and ordering work, and the time (end deadline) that must be provided as a service is determined, the destination This business computer needs to be moved to a business computer that can satisfy the above-mentioned service expiration date. Furthermore, since there is a difference in the performance of shared resources between business computers, it is difficult to determine a business computer to be a destination.

  Here, when business is outsourced at a data center, etc., when the resource usage of the service is provided to the customer or used for reviewing the virtual system configuration, etc. When there is responsibility, the amount of resources for re-execution (rerun) due to service job interruption must be excluded from aggregation. If the service user has a cause or responsibility, such as an error in the input information, the amount of resources for re-execution (rerun) must be included in the aggregation. This is not possible because the correspondence information between the cause of execution (rerun) and the re-executed (rerun) job is not held.

  Therefore, the present invention has been made in view of the above problems, and an object of the present invention is to determine a destination of a virtual machine based on the resource usage of a service in a virtual system.

In order to solve the above-described problems, the present invention provides a plurality of virtual machines that provide services on one or more business computers and can be moved on one or more business computers, and each business computer has its own business. A virtual system environment that includes a host machine that manages virtual machines in a computer, manages service configurations and shared resources of business computers used by services, and can move from the resource usage when moving virtual machines It is a service management apparatus that determines a computer and moves the virtual machine.
The service management apparatus shows service configuration information including information on the service expiration date, job management information including information on a job to be executed upon restart after each job that configures the service, and a configuration of the host machine Depending on host machine configuration information, virtual machine configuration information indicating the configuration of the virtual machine, resource management information including information on the executed job and the amount of resources used by the job, and the type of event that occurred in the virtual machine Storage unit for storing event information including coping information to be executed, and service configuration information, job management information, host machine configuration information, virtual machine configuration information, resource management stored in the storage unit when moving a virtual machine Based on information and event information, if the service running on the virtual machine is completed by the end date Cormorants condition is satisfied, and a processing unit that moves in determining the destination of the virtual machine.
Other means will be described later.

  According to the present invention, in the virtual system, it is possible to determine the destination of the virtual machine based on the resource usage of the service.

  The best mode for carrying out the present invention (hereinafter referred to as “embodiment”) will be described below with reference to the drawings (refer to drawings other than the referenced drawings as appropriate).

First, terms used in the present embodiment will be described.
A virtual machine is a logical computer that divides a physical computer by virtualization technology and operates with an independent OS, and can move between business computers.
A host machine is a machine that manages virtual machines, and is a machine that cannot be moved between business computers, such as a virtual machine. In this embodiment, one host machine is assigned to one business computer. Suppose that it is built.

  A shared resource is a logically divided resource that is shared between virtual machines on the same business computer. A plurality of logically divided resources can be assigned to a virtual machine in units of logical division. it can. On the virtual machine, even if a plurality of shared resources are allocated, they are regarded as one virtual resource. For example, in the case of a CPU (Central Processing Unit) as a shared resource, there is an LPAR (Logical Partitioning) technique.

A job is an individual work that constitutes a business, such as a command, application program, or shell script provided by a business program. Each job is the smallest unit of business processing. That is.
The job ID (IDentification) is an ID assigned to identify a collection of jobs that define a plurality of jobs as services (businesses).
An event is an event that occurs in a virtual machine. The name of the virtual machine in which the event has occurred and the content of the event that has occurred are stored inside the event.

  FIG. 1 is an overall configuration diagram of a service / resource management system according to the present embodiment. As shown in FIG. 1, the service / resource management system S (service management system) includes a service / resource monitoring device 1 and an event management device 4 (the “service management device” in the claims includes this event management device 4). .), The service management apparatus 2, and a plurality of business computers 7. The service management apparatus 2, event management apparatus 4, and business computer 7 are connected by a network 3.

  The service / resource monitoring apparatus 1 displays the shared resource configuration mounted on the business computer 7 and the information of the virtual machine 9 to which the shared resource is allocated, the virtual resource information of the virtual machine 9 in the system, the virtual machine 9 Display of configuration information of services provided above, display of configuration information of jobs provided as services, display of resource usage used by services, display of job execution history, display of event information generated in virtual machine 9 A GUI (Graphical User Interface) 10 is provided for outputting a screen for operating the handling information for display and event contents.

  The service management apparatus 2 includes a service configuration management unit 20, a virtual machine generation change unit 21, a job management unit 22, a job execution unit 25, a service configuration change unit 24, a resource information acquisition unit 23, a resource information display unit 26, and a configuration information display. The unit 27 is provided.

  The service configuration management unit 20 is a processing unit that communicates with the configuration acquisition agent unit 82 of the business computer 7 and acquires information such as the number and capacity of shared resources installed in the business computer 7. Further, the virtual machine name 682 and the host machine name 683 for creating (generating) the virtual machine name 682 are acquired from the virtual machine configuration table 55 (virtual machine configuration information) (details will be described later in FIG. 13), and the virtual machine generation change Unit 21 is requested to generate a virtual machine, and after the virtual machine is generated, an LPAR_ID 562 that can be assigned to the virtual machine in the machine configuration table 53 (host machine configuration information) (details will be described later in FIG. 11) is acquired, and the virtual machine configuration From the CPU amount 684 set in the table 55, the virtual machine name is set to the LPAR use virtual machine name 563 corresponding to the LPAR_ID 562 assigned to the virtual machine.

  The service configuration management unit 20 further determines the amount of memory and disk space required by the virtual machine from the required memory amount 516 and the required disk amount 517 in the service configuration table 51 (details will be described later in FIG. 9) (service configuration information). The amount is totaled, and the assigned virtual machine name is set in the memory ID 564 (see FIG. 11) assigned to the virtual machine, the memory use virtual machine name 567 (see FIG. 11) corresponding to the disk, and the disk use virtual machine name 570. After setting the shared resource to be allocated to the virtual machine in the machine configuration table 53, it communicates with the configuration change agent unit 83 and allocates the shared resource managed by the host machine to the virtual machine with the contents of the machine configuration table 53. After the shared resources are allocated to the virtual machine, the total of the shared resource amounts allocated to the memory amount 685 and the disk amount 686 of the virtual machine configuration table 55 is set.

  The virtual machine creation / change unit 21 is a processing unit that communicates with the virtual machine creation / change agent unit 81 of the host machine having the host machine name 683 registered in the virtual machine configuration table 55 and creates a virtual machine in the business computer 7. . Further, when moving a virtual machine, the virtual machine created on the source host machine is deleted and a virtual machine is created on the destination host machine.

  The job management unit 22 displays the information of the job management table 52 (job management information) and the job history management table 57 (job history management information) on the GUI 10 of the service / resource monitoring device 1, or from the service / resource monitoring device 1. Is a processing unit that receives information on the configuration of the job and the contents of the job to be executed. When the service start time 514 of the service configuration table 51 is reached, the job execution unit 25 is requested to execute the job, the job start time and end time and the execution result are received from the job execution unit 25, and the virtual machine configuration table 55, the virtual configuration ID 680 and the host machine name 683 corresponding to the virtual machine name that executes the job are acquired, the configuration ID 560 of the host machine name 561 is acquired from the machine configuration table 53, and the acquired information is stored in the job history management table 57. A processing unit to be registered.

  The job execution unit 25 is a processing unit that communicates with the job execution agent unit 92 of the virtual machine, sends job information to be executed, and receives a job execution result from the job execution agent unit 92.

  The service configuration changing unit 24 is a processing unit that executes the handling content received from the event handling execution unit 43. When the handling content is a virtual machine migration, the service configuration change unit 24 satisfies the resources used by the service on the migration target virtual machine. This processing unit searches for a host machine name that can be used (refer to FIGS. 18, 19A, and 19B for the processing flow). In addition, when there are multiple services in a virtual machine and there is no migration destination that satisfies the resource amount used by the entire service, there is no business computer that satisfies the resource amount by controlling the service execution order. (See FIG. 20 for the processing flow). If the destination host machine name can be acquired, the virtual machine information after the change is added to the virtual machine configuration table 55, and a virtual machine migration request is made to the virtual machine generation / change unit 21. After the movement of the virtual machine is finished, the configuration information of the source host machine and the configuration information of the destination host machine are newly added to the machine configuration table 53, and the configuration change agent of the source and destination host machines is added. It communicates with the unit 83 to notify the changed host machine configuration added to the machine configuration table 53. Further, after the movement of the virtual machine is completed, the job execution unit 25 is requested to restart the job.

The resource information acquisition unit 23 is a processing unit that communicates with the resource information acquisition agent unit 91 of the virtual machine and receives a resource amount result used for executing the job.
The resource information display unit 26 is a processing unit that displays information of the resource management table 56 (resource management information) on the GUI 10 of the service / resource monitoring apparatus 1.

  The configuration information display unit 27 displays the table information of the service configuration table 51, the machine configuration table 53, the performance coefficient table 54 (performance coefficient information), and the virtual machine configuration table 55 on the GUI 10 of the service / resource monitoring apparatus 1, It is a processing unit that receives a request from the service / resource monitoring apparatus 1.

  The storage unit 1 (5) handled by the service management apparatus 2 (note that the number in () after the name (here “storage unit 1”) (here “5”) is a sign, and so on). Table information includes a machine configuration table 53, a virtual machine configuration table 55, a service configuration table 51, a job management table 52, a resource management table 56, a performance coefficient table 54, and a job history management table 57.

The event management device 4 includes an event management unit 41, an event analysis unit 42, an event handling execution unit 43, and an event display unit 44.
The event management unit 41 is a processing unit that receives an event occurring in a virtual machine from the event notification agent unit 93 of the virtual machine and registers information stored in the event in the event management table 62 (event information).
The event analysis unit 42 is a processing unit that analyzes whether a message 611 that matches the event message 632 registered in the event management table 62 by the event management unit 41 exists in the event handling table 61 (event information).

The event handling execution unit 43 is a processing unit that notifies the service configuration changing unit 24 of the service management device 2 of the handling content 612 corresponding to the content of the event that has occurred in the virtual machine from the analysis result of the event analyzing unit 42.
The event display unit 44 displays the table information of the event handling table 61 and the event management table 62 on the GUI 10 of the service / resource monitoring apparatus 1, and the event message 611 and the handling content 612 set in the event handling table 61 from the GUI 10. It is a processing unit that accepts.

  Table information of the storage unit 2 (6) handled by the event management device 4 includes an event handling table 61 and an event management table 62.

  The business computer 7 includes a virtual machine generation change agent unit 81, a configuration acquisition agent unit 82, and a configuration change agent unit 83 in the host machine 8, and a business program 94 and resources are added to the virtual machine 9 constructed in the business computer 7. An information acquisition agent unit 91, a job execution agent unit 92, and an event notification agent unit 93 are provided.

A plurality of virtual machines 9 exist in the business computer 7. The business program 94 is an execution program for providing a service, and a plurality of business programs 94 exist in the business computer 7.
The virtual machine generation change agent unit 81 is a processing unit that creates the virtual machine 9 in the business computer 7 or deletes the virtual machine 9 from the business computer 7.
The configuration acquisition agent unit 82 is a processing unit that acquires shared resource information in the business computer 7 and notifies the service configuration management unit 20 of the service management apparatus 2 of the acquisition result.
The configuration change agent unit 83 is a processing unit that allocates shared resources to the virtual machine 9 based on the machine configuration information received from the service configuration change unit 24 of the service management apparatus 2.

The resource information acquisition agent unit 91 receives the job name and command name for acquiring the resource from the job execution agent unit 92, collects the resource usage of the received job, and collects the collection result as the resource information acquisition unit 23 of the service management apparatus 2. Notify
The job execution agent unit 92 executes the job notified from the job execution unit 25 of the service management apparatus 2 and notifies the resource information acquisition agent unit 91 of the job name and command name for acquiring the resource usage amount. The processing unit notifies the job execution unit 25 of the service management apparatus 2 of the execution result.
The event notification agent unit 93 monitors an event that occurs during execution of a job in the virtual machine. When an event occurs during execution of a job, the event notification agent unit 93 notifies the event management unit 41 of the event management device 4 of the event. Part.

  FIG. 2 is an overall configuration diagram for explaining a hardware configuration of the service / resource monitoring apparatus 1 and the service management apparatus 2 according to the present embodiment. The service management apparatus 2 includes a CPU 110 (processing unit), a memory 111 (storage unit), a storage device 112 (storage unit), a display interface 113, a disk interface 114, and a communication interface 115. The service / resource monitoring apparatus 1 includes a CPU 100, a memory 101, a storage device 102, and a display interface 103.

  The service management program 11 can be realized by the CPU 110 loading each program stored in the storage device 112 onto the memory 111 and executing it. The service management device 2 is connected to the external storage unit 1 (5) via the disk interface 114, and is connected to the event management device 4 and the business computer 7 via the communication interface 115 and the network 3. On the other hand, in the service / resource monitoring apparatus 1, the GUI 10 can be realized by the CPU 100 loading and executing the program stored in the storage device 102 onto the memory 101.

  FIG. 3 is a configuration diagram of the storage unit 1 (5). The storage unit 1 (5) stores a service configuration table 51, a job management table 52, a machine configuration table 53, a performance coefficient table 54, a virtual machine configuration table 55, a resource management table 56, and a job history management table 57.

  FIG. 4 is an overall configuration diagram for explaining the hardware configuration of the event management apparatus 4 and the business computer 7 of the present embodiment. The event management apparatus 4 includes a CPU 120 (processing unit), a memory 121 (storage unit), a storage device 122 (storage unit), a display interface 125, a disk interface 123, and a communication interface 124.

  The event management program 12 can be realized by the CPU 120 loading each program stored in the storage device 122 onto the memory 121 and executing it. The event management device 4 is connected to the external storage unit 2 (6) via the disk interface 123, and is connected to the service management device 2 and the business computer 7 via the communication interface 124 and the network 3.

  The business computer 7 is divided into logical machines of the host machine CPU 130, LPAR_1 (132) to LPAR_N (133), which can be allocated to the virtual machine, the virtual machine CPU 131, the host machine memory 137, and a plurality of virtual machines. An allocatable virtual machine memory 138, a storage device 135, and a communication interface 136 are provided. Here, the virtual machine generation change agent unit 81, the configuration acquisition agent unit 82, and the configuration change agent unit 83 load the respective programs stored in the storage device 135 by the host machine CPU 130 onto the host machine memory 137. It is realized by executing. The resource information acquisition agent unit 91, the job execution agent unit 92, and the event notification agent unit 93 are executed by the virtual machine CPU 131 loading and executing each program stored in the storage device 135 onto the virtual machine memory 138. Realized. The business computer 7 is connected to the service management device 2 and the event management device 4 via the communication interface 136 and the network 3.

FIG. 5 is a diagram illustrating a more specific example of the present embodiment.
The business system 13 includes business computers 1 (71) to M (73) (corresponding to the business computer 7 in FIG. 1). Each business computer 1 (71) to M (73) has a plurality of virtual machines A (74) to N (77) and a host machine 1 (78) that manages the virtual machines A (74) to N (77). ) To M (80). The service management apparatus 2 manages the resource information of the business computers 1 (71) to M (73) and the configuration information of the virtual machines A (74) to N (77). Further, the event management device 4 manages event information generated in each of the virtual machines A (74) to N (77). The service / resource monitoring apparatus 1 centrally monitors the business computers 71 to 73.

  The virtual memory 741, the virtual CPU 742, and the virtual disk 743, which are resources of the virtual machine A (74), include LPAR_1 (7111) and LPAR_2 (7112), memory 1 (7121), memory 2 (7122), and disk 1, respectively. (7131) and disk 2 (7132) are allocated. Also, LPAR_3 (7113), memory 3 (7123), and disk 3 (7133) are allocated to the virtual memory 751, virtual CPU 752, and virtual disk 753, which are resources of the virtual machine B (75), respectively. Note that LPAR_4 (7114), the memory 4 (7124), and the disk 4 (7134) are not allocated to the virtual machine and are surplus resources.

  A virtual machine C (76) is constructed in the business computer 2 (72). LPAR_1 (7211), Memory 1 (7221), Memory 2 (7222), and Disk 1 (7231) are allocated to the virtual memory 761, virtual CPU 762, and virtual disk 763, which are resources of the virtual machine C (76), respectively. It has been. Note that LPAR_2 (7212) to LPAR_4 (7214), the memory 3 (7223) and the memory 4 (7224), and the disk 2 (7232) to the disk 4 (7234) are not allocated to the virtual machine and become surplus resources. ing.

FIG. 6 is a diagram illustrating a service provided by the virtual machine in FIG. 5 and jobs constituting the service in the present embodiment.
A virtual machine A (74) and a virtual machine B (75) are built in the business computer 1 (71). In the virtual machine A (74), a service A1 (744) and a service A2 (745) are provided. Each of the services A1 (744) and A2 (745) includes a job ID1 (746) and a job ID2 (747) which are a collection of jobs as services. Here, the job ID 1 (746) includes jobs 1A (7461) to 1E (7465). Job 1A (7461) is a service A1 (744) start job, and when execution of job 1A (7461) is completed, execution of job 1B (7462) and job 1C (7463) is started. Other jobs are executed in the same flow.

  The job ID 2 (747) is composed of jobs 2A (7471) to 2C (7473). Further, in the virtual machine B (75), a service B (754) is provided. The service B (754) includes a job collection job ID 3 (755) as a service. Job 3A (7551) is a start job with job ID 3 (755), and when execution of job 3A (7551) is completed, execution of job 3B (7552) is started. Other jobs are executed in the same flow.

  A virtual machine C (76) is built in the business computer 2 (72). In the virtual machine C (76), a service C (764) is provided. The service C (764) includes a job ID 4 (765) that is a collection of jobs as services. Here, the job ID 4 (765) includes a job 4A (7651) and a job 4B (7652). Job 4A (7651) is a start job of service C (764), and when execution of job 4A (7651) is completed, execution of job 4B (7652) is started. The job 4B (7652) is an end job. When the job 4B (7652) ends, the provision of the service of the service C (76) ends.

  A virtual machine N (77) is built in the business computer M (73). In the virtual machine N (77), a service N (774) is provided. The service N (774) includes a job IDN (775) that is a collection of jobs as services. Here, the job IDN (775) is composed of a job NA (7751) to a job NC (7753). The flow of job execution is the same as described above.

FIG. 7 is a diagram showing an event handling table 61 which is one of the tables held in the storage unit 2 (6).
The event handling table 61 holds a relationship between a message that is a cause of an event storing failure information and the like that has occurred in the virtual machine and the handling content, and is defined (input in advance) by the operation manager.

  The handling ID 610 is an ID for identifying the message 611 and the handling content 612. The message 611 is an item for setting a message stored as an event generation factor notified from the virtual machine. The countermeasure content 612 is an item for setting content to be dealt with with respect to the message 611.

For example, when an event storing the “disk access error” message 611 of countermeasure 1 (620) is registered in the event management table 62, the event countermeasure execution unit 43 changes the disk assigned to the virtual machine “ The request “change disk” is notified to the service configuration change unit 24 of the service management apparatus 2.
If an event storing the “failure in virtual machine” message 611 of the countermeasure 2 (621) is registered, a “virtual machine move” request for moving the failed virtual machine to another business computer is issued as a service management apparatus. 2 to the service configuration change unit 24.

When an event storing the “network error” message 611 of the countermeasure 3 (622) is registered, a “virtual machine move” request for moving the affected virtual machine to another business computer is issued as a service of the service management apparatus 2. The configuration change unit 24 is notified.
When an event storing the message “CPU resource usage is 90%” in response 4 (623) is registered, an “add CPU” request to add a CPU to the corresponding virtual machine is issued as a service of the service management apparatus 2 The configuration change unit 24 is notified.

FIG. 8 is a diagram showing an event management table 62 which is one of the tables held in the storage unit 2 (6). The event management table 62 holds a relationship between an event that has occurred in each virtual machine and the action that has been taken.
The event ID 630 is assigned an ID that can be uniquely identified when registered in the event management table 62. In the virtual machine name 631, the name of the virtual machine in which the event has occurred is set. In the message 632, a message held by the event is set. The job execution ID 633 is set with the job execution ID of the job that generated the event. In the handling ID 634, the handling ID 610 (see FIG. 7) executed by the event handling table 61 that matches the message of the event set in the message 632 is set by the event analysis unit 42.

For example, when the event 1 (640) is being executed by the virtual machine A (74) and the job execution ID 633 is “1”, the event management unit 41 is notified of an event whose message 632 is “disk access error”. The action ID (634) is “action 1” (see FIG. 7), which is the action ID 610 (see FIG. 7) that matches the message 611 (see FIG. 7) in the event handling table 61. 7 (see the row of reference numeral 620).
In the countermeasure ID (634) of event 2 (641), “Countermeasure 2” (refer to the row of reference numeral 621 in FIG. 7) of the event countermeasure table 61 is set, and “Countermeasure 3” (see line 621 in FIG. 7). 622) of FIG. 7 is set, and “Countermeasure 4” (see the row indicated by reference numeral 623 in FIG. 7) is set to the event 4 (643).

FIG. 9 is a diagram showing a service configuration table 51 which is one of the tables held in the storage unit 1 (5). The service configuration table 51 holds information on services provided in the business system, and is defined (input in advance) by the operation manager.
In the service name 510, a service name provided in the business system is set. In the virtual machine name 511, a virtual machine name that provides a service is set. In the job ID 512, an ID for identifying a collection of jobs constituting the service is set (see FIG. 10). In the start job 513, a job name for starting the service is set. In the service start time 514, a time for executing the start job is set. As the service end time 515, the last job (the next job name is not set in the job management table 52 for the time when the service provision should end, that is, the job ID (512) constituting the service. The time at which the execution of the job (see FIG. 10) must end is set. In the necessary memory amount (516), a memory amount necessary for operating the service is set. In the necessary disk capacity (517), a disk free capacity necessary for operating the service is set.

For example, the service A1 (744) set in the row 520 of the service configuration table 51 is provided by the virtual machine A (74), and is composed of a job group of job ID1, and the start job (513) is the job 1A. The time for starting execution of job 1A is “00:00” every day. Further, the time when the job 1E (see FIG. 10), which is the final job of the service A1 (744), must be completed is “04:00” every day. Furthermore, in order to operate the service A1 (744), it is set that the virtual machine A (74) needs a memory capacity of 0.5 GB and a disk free capacity of 100 GB.
For other services (reference numerals 521 to 525), the virtual machine name, job ID, start job name, service start time, service end time, required memory amount, and required disk amount to be provided are set.

FIG. 10 is a diagram showing a job management table 52 which is one of the tables held in the storage unit 1 (5).
The job management table 52 holds the relationship between the job name and command name constituting the service, the job execution order, and the restart job name that starts re-execution (rerun) when job execution fails. , Defined (preliminarily input) by the operations manager.

  In the job ID (530), an ID for identifying a collection of jobs to be executed for each service is set. In the job name (531), a job name that is a component of the job ID is set. In the command name (532), the command name executed by the job with the job name is set. In the next job name (533), a job name to be executed next when the job set in the job name is completed. In the previous job name (534), the job executed before the job set in the job name is set. In the restart job name (535), a job name to be re-executed (rerun) when execution of the job set in the job name fails is set.

  For example, the job ID 1 (540 to 544) constituting the service A1 (744) provided by the virtual machine A (74) in FIG. 6 is the job 1A, job 1B, job 1C, job 1D, which is the start job of the job ID1, Consists of job 1E. When the execution of the command 1A is completed for the job 1A, the job 1B and the job 1C set in the next job name (533) are executed. If execution of job 1A fails, job 1A set in the restart job name (535) is re-executed (rerun). The job 1B is executed after the execution of the job 1A set in the previous job name (534) is completed, and when the execution of the command 1B executed in the job 1B is completed, the job set in the next job name (533) Perform 1D. If execution of job 1B fails, the job is executed again from job 1A.

The job 1C is executed after the execution of the job 1A set in the previous job name (534) is completed, and when the execution of the command 1C executed in the job 1C is completed, the job set in the next job name (533) Perform 1D. If execution of job 1C fails, the job is executed again from job 1A. The job 1D is executed after the execution of the job 1B and the job 1C set in the previous job name (534) is finished, and when the execution of the command 1D executed in the job 1D is finished, the job 1D is set in the next job name (533). Executed job 1E. If execution of job 1D fails, the job is executed again from job 1A. The job 1E is executed after the execution of the job 1D set in the previous job name (534) is finished. When the execution of the command 1E executed in the job 1E is finished, the job name is set in the next job name (533). Therefore, the execution of job ID1 is terminated, and the provision of service A1 (744) is terminated. If execution of job 1E fails, the job is executed again from job 1D.
For job ID 2 (545-547) to job ID N (554-556), the name of the job to be executed, the command name, the next job name, the previous job name, and the restart job name are set.

FIG. 11 is a diagram showing a machine configuration table which is one of the tables held in the storage unit 1 (5).
The machine configuration table 53 includes information on the virtual machine CPU 131, the virtual machine memory 138, and the storage device 135, which are shared resources installed in the business computers 1 (71) to M (73) in the business system 13. It is a table that holds the relationship between virtual machine names assigned to each shared resource, and the processing performance of each memory and disk. As the initial machine configuration, the host machine name (561) having the configuration IDs “1” (580) to “M” (593) is defined (input in advance) by the operation manager.

  An ID assigned to identify the configuration of the host machine is assigned to the configuration ID (560), and an ID is newly added every time the configuration is changed. In the host machine name (561), the host machine names of the business computers 1 (71) to M (73) managed in the business system 13 are set. LPAR_ID (562) is set with an ID for identifying LPARs by the number of LPARs managed by the machine with the host machine name. In the LPAR use virtual machine name (563), a virtual machine name using the LPAR_ID is set. In the memory ID (564), a memory ID for identifying the memory by the number of memories managed by the machine having the host machine name is set.

  In the memory capacity (565), the capacity of the memory set in the memory ID (564) is set. In the memory performance (566), the performance of the memory (information processing amount per unit time) set in the memory ID (564) is set. In the memory use virtual machine name (567), the name of the virtual machine that uses the memory set in the memory ID (564) is set. In the disk ID (568), IDs for identifying the disks are set as many as the number of disks managed by the machine having the host machine name. In the disk free capacity (569), the free capacity of the disk set in the disk ID is set. In the disk use virtual machine name (570), the virtual machine name that uses the disk set in the disk ID (568) is set. In the disk reading performance (571), the disk reading performance (reading amount per unit time) set in the disk ID is set. In the disk writing performance (572), the disk writing performance (writing amount per unit time) set in the disk ID is set.

  For example, for the host machine 1 (78) (see FIG. 5) whose configuration ID (560) is “1” (580 to 583), LPAR_1 (7111) to LPAR_4 which are logical partitioning sections of the CPU of the host machine 1 (78). (7114), the memory is composed of memory 1 (7121) to memory 4 (7124), and the disk is composed of disk 1 (7131) to disk 4 (7134). Is acquired from the configuration acquisition agent unit 82 (see FIG. 1) of the host machine 1 (78) and set in the machine configuration table 53. The operation manager defines (sets) the memory performance (566), disk read performance (571), and disk write performance (572). For the LPAR use virtual machine name (563), memory use virtual machine name (567), and disk use virtual machine name (570) that use each shared resource, the service configuration management unit 20 uses the virtual machine configuration table 55 and the service configuration. The virtual machines defined in the table 51 are automatically allocated according to the information on the necessary shared resource amount (see the description of FIG. 17).

  Here, the virtual machine A (74) (see FIG. 5) created in the host machine 1 (78) includes LPAR_1 (7111) and LPAR_2 (7112), memory 1 (7121) and memory 2 (7122), disk 1 (7131) and disk 2 (7132) are allocated. Further, the virtual machine B (75) (see FIG. 5) created in the same host machine 1 (78) as the virtual machine A (74) includes LPAR_3 (7113), memory 3 (7123), and disk 3 (7133). ) Is assigned. Note that LPAR_4 (7114), memory 4 (7124), and disk 4 (7134) are not allocated to the virtual machine.

  The configuration information (584 to 593) of the other host machines 2 (79) to M (80) also holds the configuration information of the shared resources of the host machine and the information of the virtual machines to which the shared resources are allocated. .

  When the event handling execution unit 43 requests to move the virtual machine A (74) from the host machine 1 (78), the service configuration changing unit 24 uses the amount of resources used by the service of the virtual machine A (74). The destination host machine name that satisfies the conditions and the resources to be allocated are determined (see the processing of FIGS. 19A and 19B), and the configuration (580 to 583) of the host machine 1 in the machine configuration table 53 is referred to, and the configuration ID (560 ), “M + 1” (594 to 597) is created, and the content (580 to 583) of “1” of the configuration ID (560) is copied.

  Then, in the rows 594 to 597 whose configuration ID (560) is “M + 1”, LPAR_1 (7111) and LPAR_2 (7112), memory 1 (7121), and memory 2 (7122) allocated to the virtual machine A (74). ), The used virtual machine names (563, 567, 570) of the disk 1 (7131) and the disk 2 (7132) are changed to “unused”. Further, the service configuration changing unit 24 copies “2” (584 to 587) of the configuration ID (560) of the host machine 2 (79) that is the transfer destination to “M + 2” (598 to 601), and LPAR_2 (7212) to LPAR_4 (7214), the memory 3 (7223) and the memory 4 (7224), the disk 2 (7232) and the disk 3 (7233) which are surplus resources of the machine 2 (79) are transferred to the virtual machine A. (74). The service configuration changing unit 24 uses “M + 1” (594 to 597) and “M + 2” (M + 2) (configuration IDs (560) after the change of the migration source host machine 1 (78) and the migration destination host machine 2 (79)). 598 to 601) is created and set, and then the change information is notified to the configuration change agent unit 83 of each host machine (see the process in FIG. 18).

12A is a diagram showing a performance coefficient table 54, which is one of the tables held in the storage unit 1 (5), and FIG. 12B is a graph showing an evaluation formula (details will be described later).
As shown in FIG. 12A, the performance coefficient table 54 holds the relationship between the host machine name (650) and the performance of the shared resource managed by the host machine, and is defined (input in advance) by the operation manager. is there. Note that the items to be set in the performance coefficient table are set for resources having a difference when there is a difference in the performance of the virtual machine depending on the number of shared resources allocated to the virtual machine in the host machine.

  In the host machine name (650), the host machine name in the business system 13 is set. Here, it is assumed that there is a difference in processing performance of the virtual machine depending on the number of LPARs and memory allocated to one virtual machine, and the number of instructions (average) (651) depends on the number of LPARs allocated to the virtual machine. Set the performance evaluation formula. In the memory process (652), a performance evaluation formula based on the number of memories allocated to the virtual machine is set. For disk read (653) and disk write (654), a performance evaluation formula is set according to the number of disks allocated to the virtual machine.

  For example, the number of instructions (average) of the host machine 1 (660) (651) includes the evaluation formula f1c (x) for the LPAR processing performance of the host machine 1 and the number of instructions (average) of the host machine 2 (661). Set f2c (x). Here, as shown in FIG. 12 (b), the LPAR performance evaluation formula f1c (x) of the host machine 1 (660) is 400,000 (times) / s (seconds) when the number of LPARs is one. 679), when the number of LPARs is two, 1 million / s, (677), when the number of LPARs is three, 1.4 million / s (676), and when the number of LPARs is four, The processing performance is 1.5 million / s (675). The LPAR performance evaluation formula f2c (x) of the host machine 2 has a processing performance of 800,000 / s (678) when the number of LPARs is one, and 1.9 million / s when the number of LPARs is two. When the number of s (674) LPARs is 3, the processing performance is 2 million / s (673), and when the number of LPARs is 4, the processing performance is 1.9 million / s (674). Note that, when there are a plurality of LPARs, overhead occurs, so each numerical value in the evaluation formula is as follows.

FIG. 13 is a diagram showing a virtual machine configuration table 55 that is one of the tables held in the storage unit 1 (5).
The virtual machine configuration table 55 holds the relationship between the virtual machine name (682), the host machine name (683) for constructing the virtual machine name, and the amount of shared resources allocated to the virtual machine.

  In the virtual configuration ID (680), an ID for identifying the configuration of the virtual machine is set. The virtual configuration ID (681) before change is set to the virtual configuration ID immediately before the virtual machine to be changed when the configuration of the virtual machine is changed. In the virtual machine name (682), a virtual machine name created in the business computer 7 is set. In the host machine name (683), the name of the host machine that constructs the virtual machine is set. In the CPU amount (684), the ratio (%) of the CPU resource allocated to the virtual machine is set. In the memory amount (685), a memory amount allocated to the virtual machine is set. In the disk amount (686), a disk amount allocated to the virtual machine is set.

  Note that the amount of memory allocated to the virtual machine with the virtual machine information with the virtual configuration ID (680) of “0” with the virtual configuration ID (681) before change of “1” (690) to “N” (694) ( Items other than 685) and the amount of disk (686) need to be set by the operation manager first. The memory amount (685) and the disk amount (686) are automatically set by the service configuration management unit 20 by the shared resource determination process (see the process of FIG. 17) to be allocated to the virtual machine.

When the virtual configuration ID (680) is “N + 1” (695) or later, when there is a virtual machine migration request from the event handling execution unit 43, the service configuration change unit 24 determines the host machine name of the migration destination, Determine the amount of resources to be allocated to the virtual machine to be moved on the destination host machine, and add the virtual machine information after the change.
For example, the virtual machine A (74) with the virtual configuration ID (680) of “1” (690) is created in the host machine 1 (78), and the CPU resource of the host machine 1 (78) is 50% (the number of LPARs is set). 50%), the memory amount is 1.0 GB, and the disk amount is 200 GB allocated.

For other virtual configuration IDs (680) “2” (691) to “N” (694), the creation host machine names, CPU amounts, memory amounts of virtual machines B (75) to N (77), Set the disk amount.
When the virtual configuration ID (680) is “N + 1” (695) or later, when the event handling execution unit 43 issues a migration request for the virtual machine A (74), the service configuration change unit 24 sets the migration destination. Since the host machine 2 (79) is determined as the host machine name and three LPAR_2 to LPAR_4 among the four LPARs held by the host machine 2 (79) to which the virtual host machine A (74) is moved are assigned 75 % Is set, and 1.0 GByte and 200 GByte allocated respectively for the memory amount and the disk amount are set (see the description of the processing in FIG. 18).

FIG. 14 is a diagram showing a resource management table 56 that is one of the tables held in the storage unit 1 (5).
The resource management table 56 is a table that holds the executed job and the resource amount used by the job in the virtual machine that executed the job, and is used by the resource information acquisition agent unit 91 to monitor and collect the resource usage amount. .

  The job execution ID 701 is an ID assigned every time a job is executed. The CPU usage rate (average) 702 is set to the average value of the CPU usage rate used by the executed job. In the instruction count (average) 703, an average value of the instruction count to the CPU of the executed job is set. In the memory processing 704, an average value of the memory processing amount used by the executed job is set. In the disk read 705, an average value of the amount of data read by the executed job from the disk is set. In the disk write 706, an average value of the amount of data written by the executed job to the disk is set.

  Note that the resource usage amount managed by the resource management table 56 is not limited to this, and an arbitrary item can be set. For example, a job whose job execution ID 701 is “1” uses an average of 30% CPU from the start to the end of job execution, and the number of instructions per unit time (average) is 800,000 times. And the reload amount is 64 Mbytes per unit time, and the resource information acquisition agent unit 91 monitors that the resource amount of 10 Mbytes and 10 Mbytes is used per unit time for reading and writing to the disk. The information acquisition unit 23 is notified. For other jobs whose job execution IDs 701 are “2” (711) to 112 (733), the resources used by the job, the CPU usage rate, the number of instructions (average), memory processing, disk read, and disk write are acquired.

FIG. 15 is a diagram showing a job history management table 57 which is one of the tables held in the storage unit 1 (5).
The job history management table 57 includes information on the virtual machine that executed the job, the name of the executed job, the start date and time of the job, the end date and time, and the job execution ended normally, failed, or due to a failure. It is a table that holds the relationship between whether or not the job has been re-executed (rerun) and the event ID (630) (see FIG. 8) of the event management table 62 that caused the job to be re-executed.

  An ID for identifying the executed job is assigned to the job execution ID (801) at the time of job execution. In the job ID (802), the job ID of the executed job is set. In the virtual configuration ID (803), a virtual configuration ID for specifying virtual machine information when the job is executed is set. Similarly, the configuration ID (804) is set with a configuration ID for specifying the host machine information. In the job name (805), the name of the executed job is set. In the job start date and time (806), the date and time when the job was started is set. In the job end date and time (807), the date and time when the job is ended is set. In the type (808), a factor for executing the job is set. When N is set, it indicates that the job is started at the service start time in the service configuration table. Further, when E is set, it indicates that a failure has occurred during execution of the job, and when R is set, it indicates that execution of the job has failed and re-executed (rerun). In the event ID (809), the event ID that causes the job with the job execution ID to be re-executed is set.

  For example, the job with the job execution ID (801) “1” is the job 1A with the job ID 1, the virtual configuration ID (803) at the time of job execution is “1” (690 in FIG. 13), and the configuration ID (804). Is “1” (580 in FIG. 11). Further, the date and time when the job was started is “200x / 01/01 0:00”, the date and time when the job is finished is “200x / 01/01 0:15”, and the job has been completed normally. Therefore, N is set as the type.

  Further, since the job having the job execution ID (801) “103” (824) has failed to be executed, E is set as the type. Further, the job with the job execution ID (801) “105” (826) has failed to execute the job with the job execution ID “103” (826), so the job 1A, which is the re-execution job of the job 1D, is re-executed. Since it has been executed (rerun), R is set as the type. Here, when the execution of the job having the job execution ID (801) of “103” fails, the event 2 (641 in FIG. 8) occurs, and the virtual machine A moves from the host machine 1 to the host machine 2. In the virtual configuration ID (803) and the configuration ID (804), “N + 1” (695 in FIG. 13) and “M + 2” (599 to 601 in FIG. 11), which are the configuration information after the change, are set. In the event ID (809), “2” (event 2) (641 in FIG. 8) is set as the event ID that has caused the job to be executed.

FIG. 16 is a diagram showing the contents of processing of the service configuration management unit 20 that acquires the number and capacity of shared resources installed in the host machine set in the machine configuration table 53.
First, in step S161, a list of host machine names in the business system 13 is acquired from the machine configuration table 53 (see FIG. 11), and the process proceeds to step S162. As an example, the host machine name (561) in the machine configuration table 53 of FIG. 11 acquires the host machine 1 (580) to the host machine M (590).

In step S162, the host machine name for acquiring configuration information is acquired from the list of host machine names acquired in step S161, and the process proceeds to step S163.
In step S163, a configuration acquisition request is notified to the configuration acquisition agent unit 82 of the host machine name acquired in step S162, and the number and capacity of shared resources installed in the host machine are acquired. As an example, for the host machine 1 (78) in the machine configuration table 53 of FIG. 11, the number of LPARs is four, LPAR_1 (7111) to LPAR_4 (7114) are acquired, the number of memories is four, and the memory 1 (7121) to memory 4 (7124) are acquired, and each memory capacity is acquired as 0.5 GB, and the number of disks is 4 and the free capacity of each disk is 100 GB. .

In step S164, the configuration information acquired in step S163 is registered in the machine configuration table 53 (see FIG. 11), and the process proceeds to step S165. As an example, the information acquired in step S163 is registered in the LPAR_ID, the memory ID, the memory capacity, the disk ID, and the free disk capacity in FIG.
In step S165, if configuration information has been acquired in step S163 for all host machine names in the host machine name list acquired in step S161 (Yes), the process ends. Otherwise (No), the process returns to step S162. As an example, configuration information is also acquired for the host machine 2 (79) to the host machine M (80) in FIG.

FIG. 17 is a diagram illustrating processing of the service configuration management unit 20 that creates a virtual machine from the virtual machine configuration table 55 and assigns a shared resource that can be assigned to the virtual machine to the created virtual machine.
First, in step S171, a virtual machine name list is acquired from the virtual machine configuration table 55, and the process proceeds to step S172. For example, the virtual machines A (74) to N (77) are acquired from the virtual machine configuration table 55 of FIG.

In step S172, a virtual machine name is acquired from the virtual machine name list acquired in step S171, and the process proceeds to step S173. For example, the virtual machine A (74) is acquired.
In step S173, the host machine name for creating the virtual machine name acquired in step S172 is acquired from the host machine name (683 in FIG. 13) in the virtual machine configuration table 55, and the virtual machine name and host machine name are generated as a virtual machine. The change unit 21 is notified and a virtual host generation request is made. When the virtual machine creation / change unit 21 creates a virtual machine, the process proceeds to step S174. For example, the host machine 1 (78) is acquired as the name of the host machine that creates the virtual machine A (74).

In step S174, a list of services provided by the virtual machine name acquired in step S172 from the service configuration table 51 is acquired from the service configuration table 51, and the process proceeds to step S175. For example, the service A1 (744) and the service A2 (745) are acquired as services provided by the virtual machine A (74).
In step S175, the memory amount and disk free space required for the service name acquired in step S174 are totaled from the required memory amount (516) and the required disk amount (517) in the service configuration table 51 (see FIG. 9). For example, the memory amount and the disk capacity that need to be allocated to the virtual machine A (74) are the sum of the required memory amount and the disk free capacity of the service A1 (744) and the service A2 (745), and the memory amount is 1.0 GB. The disk amount is 200 GBytes.

In step S176, the host machine name (683) and CPU amount (684) corresponding to the virtual machine name (682) to be created are acquired from the virtual machine configuration table 55 (see FIG. 13), and the process proceeds to step S177. For example, in the virtual machine A (74), the host machine name to be created is the host machine 1 (78), and the CPU amount (684) allocated to the virtual machine A (74) is the LPAR_ID held by the host machine 1 (78). 50% of the number.
In step S177, the number of LPAR_IDs and the number of unassigned LPAR_IDs held by the host machine name that creates the virtual machine acquired in step S176 is acquired from the machine configuration table 53, and the process proceeds to step S178. For example, in the case of FIG. 11, the number of LPAR_IDs of the host machine 1 (78) is four, and all are unassigned.

In step S178, from the number of unallocated LPAR_ID and the number of LPAR_ID acquired in step S177 and the amount of CPU allocated to the virtual machine acquired in step S176 (684), the “number of LPAR_ID” * “CPU usage (usage) It is determined whether the ratio is equal to or less than the number of unassigned LPAR_IDs. If the condition is satisfied (Yes), the process proceeds to step S179. If the condition is not satisfied (No), the process proceeds to step S189. For example, in the case of FIG. 11, the host machine 1 (78) has the condition that the number of LPAR_IDs (4) * CPU usage (0.5) = 2 and the number of currently unassigned LPAR_IDs is “4”. Meet.
In step S179, an unallocated memory ID held by the name of the host machine that creates the virtual machine acquired in step S176 is acquired from the machine configuration table 53, and the process proceeds to step S180. For example, in the case of FIG. 11, all the memory IDs of the host machine 1 (78) are unassigned.

  In step S180, from the total capacity of the unallocated memory ID acquired in step S179 and the memory amount necessary for the virtual machine acquired in step S175, whether the memory amount necessary for the virtual machine is equal to or less than the total amount of unallocated memory. If it is determined and the condition is satisfied (Yes), the process proceeds to step S181. If the condition is not satisfied (No), the process proceeds to step S189. For example, in the case of FIG. 11, the host machine 1 (78) satisfies the condition because all the memories are unallocated and the amount of memory required for the virtual machine A (74) is 1 GByte.

In step S181, an unassigned disk ID held by the name of the host machine that creates the virtual machine acquired in step S176 is acquired from the machine configuration table 53, and the process proceeds to step S182. For example, in the case of FIG. 11, all the disk IDs of the host machine 1 are unassigned.
In step S182, the amount of disk required for the virtual machine is less than or equal to the total amount of unallocated disk, based on the total capacity of the unallocated disk ID acquired in step S181 and the amount of disk required for the virtual machine acquired in step S175. If the condition is satisfied (Yes), the process proceeds to step S183. If the condition is not satisfied (No), the process proceeds to step S189. For example, in the case of FIG. 11, the host machine 1 (78) satisfies the condition because all the disks are unallocated and 400 Gbytes and the virtual machine A (74) requires 200 Gbytes.

  In step S183, virtual machine names to be assigned are set to unassigned LPAR_IDs in the machine configuration table 53 by the number of LPAR_IDs used by the virtual machines obtained in step S178, and the process proceeds to step S184. For example, in the case of FIG. 11, the virtual machine A is set to the LPAR use virtual machine name 563 whose LPAR_ID (562) of the host machine 1 is LPAR_1 (580) and LPAR_2 (581).

  In step S184, a virtual machine name to be assigned is set to an unallocated memory ID in the machine configuration table 53 by the amount of memory used by the virtual machine obtained in step S180, and the process proceeds to step S185. For example, in the case of FIG. 11, the virtual machine A is set in the memory use virtual machine name 567 of the memory 1 (580) and the memory 2 (581) as the memory ID of the host machine 1.

  In step S185, the virtual machine name to be assigned is set to the unallocated disk ID in the machine configuration table by the amount of disk used by the virtual machine obtained in step S182, and the process proceeds to step S186. For example, in the case of FIG. 11, the virtual machine A is set in the disk use virtual machine names (570) of the disk 1 (580) and the disk 2 (581) of the host machine 1.

  In step S186, the resource allocation information to the virtual machine set in the machine configuration table 53 in steps S183 to S184 is notified to the configuration change agent unit 83 of the host machine that creates the virtual machine, and the shared resources of the host machine are notified. Assign to a virtual machine. When the allocation of the shared resource by the configuration change agent unit 83 ends, the process proceeds to step S187. For example, in the case of FIG. 11, the LPARs assigned to the virtual machine A (74) are LPAR_1 (580) and LPAR_2 (581), the memory is the memory 1 (580) and the memory 2 (581), and the disk 1 is the disk 1 (580). And disk 2 (581).

  In step S187, in the virtual machine configuration table 55, the memory amount allocated to the virtual machine acquired in step S172 in steps S183 to S185 is set to the memory amount 685, the disk amount is set to the disk amount 686, and the process proceeds to step S188. . For example, in the case of the virtual machine A (690), 1.0 GB is set as the memory amount and 200 GB is set as the disk amount.

In step S188, if all the virtual machines in the virtual machine name list acquired in step S171 are created and shared resources are assigned (Yes), the process ends. When that is not right (No), it returns to step S172.
Step S189 is a step when the conditions are not satisfied in steps S178, 180, and 182. The CPU amount 684 set in the virtual machine configuration table 55 by the operation administrator or the necessary memory amount set in the service configuration table 51 516 and the necessary disk capacity 517 are notified that the virtual machine cannot be allocated, and the process is terminated. The operation administrator who has received the notification changes the host machine that creates the virtual machine or changes the service configuration.

FIG. 18 is a diagram illustrating processing of the service configuration changing unit 24 that has received a virtual machine migration request from the event handling execution unit 43.
First, step S1801 is a process of determining a migration destination host machine (refer to FIGS. 19A and 19B for the contents of the process). When the search of the migration destination host machine is completed, the process proceeds to step S1802. For example, a failure occurs while job 1C (7463) with job ID 1 (746) is being executed on virtual machine A (74), and event 2 (641) of event management table 62 is transferred from event notification agent unit 93 to event management unit 41. ) And the event handling execution unit 43 requests the service configuration changing unit 24 to execute “virtual machine migration” of the handling 2, it is assumed that the host machine 2 (79) is determined as the migration destination of the virtual machine.

If it is determined in step S1802 that the destination host machine has been determined in step S1801 (Yes), the process advances to step S1803. Otherwise (No), the process proceeds to step S1823.
In step S1803, a new configuration ID (row) is added to the configuration ID (560) in the machine configuration table 53, the configuration information of the migration source host machine is copied to the item of the added configuration ID, and the process proceeds to step S1804. For example, “M + 1” (583) is added to the configuration ID (560), and the contents of the configuration information of the host machine 1 (78) that is the migration source host machine with the configuration ID “1” (580 to 583) are copied. To do.

  In step S1804, for the configuration ID (row) newly added to the machine configuration table 53 in step S1803, the LPAR use virtual machine name 563 and the memory use virtual machine name 567 of the shared resource allocated to the virtual machine name to be moved The disk use virtual machine name 570 is changed to unused, and the process advances to step S1805. For example, with respect to the row whose configuration ID is “M + 1” (594 to 597), the virtual machines used for LPAR_1 and LPAR_2, memory 1 and memory 2, and disk 1 and disk 2 assigned to the moving virtual machine A (74) The name is changed from virtual machine A to unused.

  In step S1805, a new configuration ID (row) is added to the configuration ID (560) in the machine configuration table 53, and the configuration information of the destination host machine name determined in step S1801 is copied to the added configuration ID item. The process proceeds to step S1806. For example, when “M + 2” (587) is added to the configuration ID and the virtual machine A (74) is moved to the host machine 2 (79) in step S1801, the configuration ID is “2” (574 to 578). Copy the contents of the configuration information of the destination host machine.

In step S1806, for the configuration ID (row) added in step S1805, the virtual machine name to be moved is set to the used machine name of the shared resource of the migration destination host machine. Note that the shared resource assigned to the virtual machine name to be moved is acquired in steps S1913 to S1920 in FIG. 19B.
For example, the virtual machine A is set to the LPAR use virtual machine name (563) whose LPAR_ID (562) is LPAR_2 (599) to LPAR4 (601) whose configuration ID is “M + 2” (598 to 601) in the machine configuration table 53 Then, the virtual machine A is set in the memory use virtual machine name (567) of the memory 2 (599) and the memory 3 (600) as the memory ID (564), and the disk ID (568) as the disk 2 (599). Virtual machine A is set in the disk use virtual machine name (570) of disk 3 (600).

  In step S1807, the configuration information added in step S1804 is notified to the configuration change agent unit 83 of the source host machine, the configuration change of the source host machine is changed, and the process proceeds to step S1808. For example, the configuration information of the host machine 1 is changed to a configuration in which the configuration ID of the machine configuration table is “M + 1” (594 to 597).

In step S1808, the configuration information added in step S1806 is notified to the configuration change agent unit 83 of the destination host machine, the configuration of the destination host machine is changed, and the process proceeds to step S1809.
For example, the configuration information of the host machine 2 is changed to a configuration in which the configuration ID of the machine configuration table is “M + 2” (598 to 601).

  In step S1809, a new virtual configuration ID is added to the virtual configuration ID (680) of the virtual machine name that has been dealt with in the virtual machine configuration table 55, and the pre-change virtual ID (681) is the target of change. The latest virtual configuration ID of the virtual machine is set, the virtual machine name to be moved is set as the virtual machine name, and the destination host machine name is set as the host machine name. The CPU amount (684), the memory amount (685), and the disk amount (686) are set by summing up the resource amount allocated to the moved virtual machine from the machine configuration table 53. After setting each value, the process proceeds to step S1810. For example, “1” is set in the pre-change virtual configuration ID (681) whose virtual configuration ID (680) is “N + 1” (695), and the virtual machine A and host machine name ( In 683), the host machine 2 is set. The CPU amount (684) is set to 75% because three of the four LPARs are allocated, and the memory amount (685) is set to 1.0 GB, and the disk amount (686) is set to 200 GB. .

  In step S1810, in the process of FIG. 20, the destination host machine that satisfies the resource amount could be determined by changing the service execution order so that the next service is executed after one service is completed. If yes (Yes), the process proceeds to step S1811. When that is not right (No), it progresses to step S1818 (refer description of the process of FIG. 20). For example, for the service A1 (744) and the service A2 (745) provided on the virtual machine A (74), the service A1 (744) and the service A2 (745) are re-transferred at the destination of the virtual machine A (74). If the resource amount can be satisfied even if the start job is executed at the same time, the restart jobs for the service A1 (744) and the service A2 (745) are executed simultaneously (proceed to step S1818). For the restart jobs of service A1 (744) and service A2 (745), the restart job of service A1 (744) is executed, and the restart job of service A2 (745) is executed after the job of the final job is completed. If the resource amount can be satisfied, the process proceeds to step S1811.

  In step S1811, the job execution unit 25 is requested to execute the restart job of the service whose service end time is the earliest compared to other services in step S2001 of FIG. 20, and when the job execution ends, the process proceeds to step S1812. move on. For example, in the service A1 (744) and the service A2 (745) in the service configuration table 51, since the end time of the service A1 (744) is earlier, the job 1A (7461) that is a restart job of the service A1 (744) Is notified to the job execution unit 25.

  In step S1812, the next job name (533) to be executed next is acquired from the job management table 52, and the process advances to step S1813. For example, when job 1A (540) that is the restart job name (535) of job 1D (544) is executed, job 1B (7462) and job 1C (7463) are acquired from the next job name (533). To do.

In step S1813, if the next job name (533) does not exist (Yes) in step S1812, the process proceeds to step S1815. If it exists (No), the process proceeds to step S1814.
In step S1814, the acquired job is executed, and the process returns to step S1812.
In step S1815, a service (next service) whose service execution order is shifted is acquired, and the process advances to step S1816. For example, when a restart job for the service A2 (745) is executed after the service A1 (744), the service A2 (745) is acquired.

  If it is determined in step S1816 that a restart job has not yet been executed and there is a service (service can be acquired) (Yes), the process advances to step S1817. If it does not exist (service could not be acquired) (No), the process ends. For example, when execution of job 2C (7473), which is the final job of service 2A (745), is completed, there is no service to be started. When the execution of job 1E (7464), which is the final job of service A1 (744), is completed, service A2 (745) exists as a service to be started.

In step S1817, the job execution unit 25 is notified of the restart service for the service next to the service executed in step S1811, and when the job execution is completed, the process returns to step S1812. For example, job 2A (7471), which is a restart job of service A2 (745) executed after service A1 (744) ends, is executed.
In step S1818, all restart jobs that must be re-executed as the virtual machine moves are notified to the job execution unit 25, and when the job execution is completed, the process proceeds to step S1819. For example, job 1A (7461) that is a restart job of service A1 (744) and job 2A (7471) that is a restart job of service A2 (745) are executed simultaneously.

In step S1819, the job name of the next job name (533) is acquired from the job management table 52.
In step S1820, if the next job name exists in the acquisition of the next job name in step S1819 (No), the process returns to step S1819. If it does not exist (Yes), the process proceeds to step S1821.

In step S1821, it is determined from the job end date / time (807) in the job history management table 57 whether there is a service job being restarted that is being executed. If there is an unfinished job (No), the process returns to step S1819. If the final job of the service has ended (Yes), the process ends.
In step S1823, if the destination host machine cannot be determined in step S1802, the operation manager is notified that the destination has not been determined.

FIG. 19A and FIG. 19B show the process of determining the destination host machine name in the process of the service configuration changing unit 24 that has received the virtual machine migration request from the event handling execution unit 43 (the process of step S1801 in FIG. 18). FIG.
In step S1901, the service name (510) and job ID (512) provided on the moving virtual machine are acquired from the service configuration table 51, and the process advances to step S1902. For example, when moving the virtual machine A (74), the service names are acquired as service A1 (744) and service A2 (745), and the job ID is acquired as job ID1 (746) and job ID2 (747).

  In step S1902, a list of job execution IDs (801) of the latest job ID (801) that matches the job ID (801) acquired in step S1901 is acquired from the job history management table 57, and the process proceeds to step S1903. For example, when job ID 1 (746) and job ID 2 (747) are acquired in step S1902, the latest job execution before moving the virtual machine A (74) from the job history management table 57 and re-executing the job is executed. In the ID list, the job execution ID (801) acquires “100” (821) to “104” (825).

  In step S1903, if there is a job execution ID (800) that has not ended at the job end date and time (807) of the job history management table 57 corresponding to the job execution ID list acquired in step S1902, (Yes), step S1904 is executed. Proceed to If all jobs with the job execution ID have been completed (No), the process advances to step S1905.

In step S1905, when the type (808) of the job history management table 57 corresponding to the job execution ID list acquired in step S1902 is set to E indicating that an error has occurred and execution has failed (No). The process proceeds to step S1904. Otherwise (Yes), the process proceeds to step S1906.
In step S1904, in steps S1903 and S1905, the job name of the job execution ID that has not ended and the job execution ID that ended in error are stored in the failed job list, and the process advances to step S1906. For example, if the execution IDs “100” (821) to “104” (825) are acquired in step S1902, the job 1D (824) and the job 2A (825) are included in the failed job list. Stored.
If it is determined in step S1906 that all job execution IDs in the job execution ID list acquired in step S1902 have been determined in step S1903 (No), the process advances to step S1907. When that is not right (Yes), it returns to step S1903.

  In step S1907, the restart job name (535) of the job stored in the failed job list is acquired from the job management table 52, and stored in the restart job list. For example, in the case of FIG. 10, the job 1A (543) that is the restart job name (535) of the job 1D and the job 2A (545) that is the job of the restart job name (535) of the job 2A are acquired.

  In step S1908, the job name to be executed after moving the virtual machine including the restart job acquired in step S1907 is acquired from the job management table 52, and the job to be executed after moving the virtual machine is acquired from the job history management table 57. The latest job execution ID (801) that has ended normally by name is acquired. For example, job ID1 and job ID2 are acquired as job IDs in step S1901, the restart jobs acquired in step S1907 are job 1A and job 2A, respectively, and the job names to be executed after moving the virtual machine are job 1A to job 1E. The latest job execution IDs of jobs 2A to 2C that have ended normally are “100” (821) for job 1A, “101” (822) for job 1B, “102” (823) for job 1C, and job 1D. Is "4" (813), job 1E is "5" (814), job 2A is "6" (815), job 2B is "7" (816), job 2C is "8" (817) Obtained as a list of execution IDs.

  In step S1909, the average resource usage of the job execution ID in the latest job execution ID list acquired in step S1908 is acquired from the resource management table 56, and the execution time is obtained from the start and end times of each job. Calculate the total amount of resources used for execution and aggregate each service. When this counting process is completed, the process proceeds to step S1910. For example, the total amount of resources used for the service A1 (744) is 603 million instructions, the memory processing amount is 864000 Mbytes, the disk read is 72000 Mbytes, and the disk write is 72000 MBytes. The total amount of resources used for the service A2 (745) is 162.0 million instructions, the memory processing amount is 1843200 Mbytes, the disk read is 144000 Mbytes, and the disk write is 144000 MBytes.

  In step S1910, the time at which the service provided by the moving virtual machine must end (end deadline) is acquired from the service end time 515 from the service configuration table 51, and the process proceeds to step S1911. For example, the end time limit (service end time 515) of the service A1 (744) is “04:00”, and the end time limit (service end time 515) of the service A2 (745) is “08: 0”.

  In step S1911, the memory amount (necessary memory amount 516) and the disk amount (necessary disk amount 517) required for the virtual machine to be moved are acquired from the service configuration table 51, and the process proceeds to step S1912. For example, the amount of memory required for the virtual machine A (74) is 1.0 GByte, and the required amount of disk is 200 GByte.

  In step S1912, a host machine name list other than the host machine name in which the virtual machine to be moved is created is acquired from the machine configuration table 53, and the process proceeds to step S1913. For example, in the case of FIG. 11, host machines 2 (79) to M (80) are acquired as host machine names other than the host machine 1 (78) in which the virtual machine A (74) is created.

  In step S1913, if the unallocated memory amount of the host machine name acquired in step S1912 is larger than the memory amount necessary for the virtual machine to be moved acquired in step S1911 from the machine configuration table 53 (Yes), step S1915. Proceed to When that is not right (No), it progresses to step S1914. For example, the unallocated memory amount (565) of the host machine 2 (79) is 1.0 GB of the memory 3 (586) and the memory 4 (587), and the virtual machine A (74) set in the service configuration table 51 ) Satisfies the condition of the amount of memory (516) required.

  In step S1915, if the unallocated disk size of the host machine name acquired in step S1912 is larger than the disk size required for the virtual machine to be moved acquired in step S1911 from the machine configuration table 53 (Yes), step S1916 is executed. Proceed to When that is not right (No), it progresses to step S1914. For example, the unallocated disk free capacity (569) of the host machine 2 (79) is 300 GB of the disk 2 (585) to the disk 4 (587), and the disk required for the virtual machine A (74) in the service configuration table 51 Satisfy the condition of quantity (517).

In step S1916, the host machine name that can provide the memory amount and the disk amount necessary for the virtual machine to be moved is stored in the migration candidate host machine name list, and the process advances to step S1914. For example, it is assumed that host machine 2 (79) and host machine M (80) are stored in the migration candidate host machine name list.
In step S1914, if a host name that has not executed step S1913 can be acquired from the host machine name list acquired in step S1912 (Yes), the process proceeds to step S1913. Otherwise (No), the process proceeds to step S1917.

In step S1917, if no host machine name is stored in the list of migration candidate host machine names acquired in step S1916 (Yes), the process ends. When that is not right (No), it progresses to step S1918.
In step S1918, the total resource usage for each service acquired in step S1909 is divided by the time from the current time to the end time (end deadline) for each service acquired in step S1910 to calculate the required resource average amount. For example, if the time from the current time to the end date of service A1 is 2 hours, the required resource average amount for service A1 (744) is 840,000 instructions / s, the memory processing amount is 120 Mbytes / s, and the disk Reading is 10 MByte / s, and disk writing is 10 MByte / s. Also for the service A2, the required resource average amount is 750,000 instructions / s for the number of instructions, 86 Mbyte / s for the memory processing amount, 7 Mbyte / s for disk reading, and 7 MByte for disk writing.

In step S1919, the host machine name stored in step S1916 is acquired from the migration candidate host machine list, and the process proceeds to step S1920. For example, the host machine 2 is acquired.
In step S1920, the performance evaluation formula for the shared resource with the host machine name acquired in step S1916 is acquired from the performance coefficient table 54, and the performance of each resource is calculated (derived) from the number of shared resources that can be allocated to the host machine. Then, the process proceeds to step S1921. For example, since the evaluation expression of the number of instructions (average) of the host machine 2 is f2c (x) and three LPAR_IDs of the host machine 2 (79) are not assigned, three LPAR_IDs are assigned from f2c (x). In this case, the number of instructions (average) is 2 million times / s (673). Here, the performance of memory processing, disk reading, and disk writing is also calculated using 652 to 654 of the performance coefficient table of the host machine 2 and is set to 256 MByte, 20 MByte, and 20 MByte, respectively.

  In step S1921, the performance value of each resource of the migration candidate destination host machine name acquired in step S1920 is the resource average amount required for the entire service (virtual machine unit) acquired in step S1918 for all resource items. When it exceeds (Yes), it progresses to step S1925. If even one resource item is not exceeded, the process proceeds to step S1922. For example, the performance of each resource of the host machine 2 (79) acquired in step S1920 is as follows: the number of instructions (average) is 2 million times / s (673), the memory processing is 256 Mbytes, the disk read is 20 Mbytes, and the disk write is 20 Mbytes. is there. The resource performances required for the virtual machine A (74) acquired in step S1918 are: the number of instructions (average) is 1.6 million times / s, the memory processing is 206 Mbyte / s, the disk read is 17 Mbyte / s, and the disk write is 17 Mbyte / s. s, which exceeds all resource items.

In step S1922, if the host machine name that becomes the next migration destination candidate can be obtained from the migration candidate destination host machine name list obtained in step S1916 (Yes), the process returns to step S1919. Otherwise (No), the process proceeds to step S1923.
In step S1923, whether or not the host machine that has become a migration candidate can satisfy the resource amount required for each service by executing the service one by one after the execution of the service has been completed. It is a process to judge. When the resource amount check process by the execution order control of the service ends, the process proceeds to step S1924.

In step S1924, if the destination host machine can be acquired by controlling the service execution order as the processing result of step S1923 (Yes), the process proceeds to step S1925. If not (No), the process is terminated.
In step S1925, the migration candidate machine that is the target of processing in step S1921 or step S1924 is determined (set) as the migration destination host machine, and the process ends.

  FIG. 20 is a diagram illustrating processing for determining a destination host machine by changing the execution order of services on a virtual machine. FIG. 20 shows the process of step S1923 in FIG. 19B. When the service restart job on the virtual machine is executed simultaneously, the condition of step S1921 in FIG. 19B is not satisfied, but the execution order is controlled for each service. Thus, it is a process of checking whether the resource amount is satisfied. In the following processing, it is assumed that the number of instructions (average) in the host machine 2 acquired from the performance coefficient table 54 in step S1920 in FIG. 19B is 1.4 million times / s.

In step S2001, the service name with the earliest service end time (515) is acquired from the service configuration table 51 among the services provided on the moving virtual machine, and the process advances to step S2002. For example, in the service A1 (744) and the service A2 (745), the service end time is earlier in the service A1 (744).
In step S2002, the host machine name is acquired from the migration candidate host machine list acquired in step S1916 of FIG. 19B, and the process proceeds to step S2003. For example, assume that the host machine 2 (79) is acquired.

  In step S2003, the resource performance value of the host machine acquired in step S1920 in FIG. 19B exceeds the resource amount acquired in step S2001 calculated in step S1918 in FIG. 19B or the resource amount recalculated in step S2007. If yes (Yes), the process proceeds to step S2004. If not (No), the process proceeds to step S2008. The required resource average amount of the service A1 (744) is 840,000 instructions / s, the memory processing amount is 120 MByte / s, the disk read is 10 MByte / s, and the disk write is 10 MByte / s. Here, when only the service A1 (744) is re-executed, the resource performance of the host machine 2 (79) can be satisfied. In step S2007, the resource average amount required by the recalculated service A2 (745) is 1.13 million times / s for the number of instructions, 128Mbyte / s for the memory processing amount, 10Mbyte / s for reading the disk, and writing to the disk. The service A2 (745) can be satisfied with the resource performance of the host machine 2 (79).

In step S2004, if all the services provided on the virtual machine can satisfy the required resource amount (Yes), the process proceeds to step S2009. Otherwise (No), the process proceeds to step S2005.
In step S2005, the service having the next service end time (515) next is acquired from the service configuration table 51, and the process proceeds to step S2006. For example, the service with the earliest service end time (515) after the service A1 (744) of the virtual machine A (74) is the service A2 (745).
In step S2006, the service start time is set to the time at which the previous service must end.

  In step S2007, the total amount of resources required for the service acquired in step S1909 in FIG. 19B is divided by the time from the previous service end time (515) to the service end time acquired in step S2005, and the required resource average amount The process returns to step S2003 via step S2008 Yes and step S2002. For example, the total amount of resources required for the service A2 (745) acquired in step S1920 in FIG. 19B is 162.0 million instructions, memory performance is 1843200 Mbytes, disk read is 144000 Mbytes, and disk write is 144000 Mbytes. When the value is divided by 4 hours from “04:00” at the time when the service A1 (744) ends to “08:00” at the time when the service A2 (745) ends, the necessary service A2 (745) is required. The average resource amount is 1.13 million times / s, the memory performance is 128 Mbyte / s, the disk read is 10 MByte / s, and the disk write is 10 MByte / s.

In step S2008, if the next host machine can be acquired from the migration candidate host machine list (Yes), the process returns to step S2002, and if not (No), the process ends.
In step S2009, by controlling the service execution order, the name of the migration candidate host machine that can satisfy the resource amount is set in the migration destination host machine, and the process ends.

FIG. 21 is a diagram showing job execution processing in step S1814 of FIG. The job execution is processed by the job execution unit 25. The job execution unit 25 receives a job name to be executed and a job ID.
In step S2101, the virtual machine name (511) that executes the job corresponding to the received job ID is acquired from the service configuration table 51, and the process proceeds to step S2102. For example, when executing a job with job ID 1 and job name job 1 A, the virtual machine A (74) corresponding to job ID 1 (520) in the service configuration table 51 is acquired.

  In step S2102, a job execution ID is newly added to the job execution ID (801) of the job history management table 57, the job ID to be executed is set in the job ID (802), and the job name to be executed in the job name (805). Is set (registered), and the process proceeds to step S2103. For example, if the job ID is job ID 1, the job name to be executed is job 1A, and the job execution ID is “1” (810), job 1D1 is set in job ID (802), and job name is set in job name (805). Set 1A.

  In step S2103, the job execution agent unit 92 having the virtual machine name acquired in step S2101 is notified of the name of the job to be executed and the command name (532) corresponding to the job name (531) of the job management table 52. When the job execution agent unit 92 finishes executing the job, the process advances to step S2104. For example, the job execution agent unit 92 is notified of the job 1A and the command 1A corresponding to the job 1A.

  In step S2104, the start date / time and end date / time of the job received from the job execution agent unit 92 in step S2103 are set in the job start date / time (806) and job end date / time (807) of the job history management table 57, and the process proceeds to step S2105. move on. For example, when the job having the job execution ID “1” (810) is executed, “200x / 01/01 00:00” is set as the job start date and time (806), and “200x / 01 / is set as the job end date and time (807). “01 00:15” is set.

  In step S2105, the virtual configuration ID (680) of the virtual machine that executed the job is acquired from the virtual machine configuration table 55, and the process advances to step S2106. For example, when the job having the job execution ID “1” (810) is executed, “1” (690) that is the latest virtual configuration ID of the virtual machine name acquired in step S2101 is acquired from the virtual machine configuration table 55. To do.

  In step S2106, the host machine name (683) corresponding to the virtual configuration ID acquired in step S2105 is acquired from the virtual machine configuration table 55, and the latest configuration ID (560) of the acquired host machine name is acquired from the machine configuration table 53. ) And the process proceeds to step S2107. For example, when “1” (690), which is the virtual configuration ID, is acquired in step S2105, from the virtual machine configuration table 55. The host machine 1 is acquired, and “1” (580) which is the latest configuration ID of the host machine 1 is acquired from the machine configuration table 53.

In step S2107, the virtual configuration ID acquired in step S2105 and the configuration ID acquired in step S2106 are converted into a virtual configuration ID (803) and a configuration ID (804) corresponding to the execution ID of the job executed in the job history management table 57. Set, and proceed to step S2108.
In step S2108, if the job execution result notified to the job execution agent unit 92 in step S2103 is normally completed (Yes), the process advances to step S2109. When the process ends abnormally (No), the process proceeds to step S2110.

In step S <b> 2109, when the job execution unit 25 accepts job execution and the notified job execution type is re-execution (rerun) (Yes), the process advances to step S <b> 2111. In other cases (No), the process proceeds to step S2112.
In step S2111, “R” indicating re-execution (rerun) is set (registered) in the type (808) corresponding to the executed job execution ID (801) in the job history management table 57, and the process proceeds to step S2113.

In step S2112, “N” indicating normal execution is set in the type (808) corresponding to the executed job execution ID (801) in the job history management table 57, and the process proceeds to step S 2113.
In step S2110, “E” indicating job execution failure is set in the type (808) corresponding to the job execution ID (801) executed in the job history management table, and the process advances to step S2113.
In step S2113, when the resource information acquisition unit 23 is requested to acquire the resource amount used by the job and the resource amount used by the executed job is received from the resource information acquisition unit 23, the process proceeds to step S2114.

  In step S2114, the job execution ID (701) of the job executed in the resource management table 56 is set, and the resource (use) amount acquired in step S2113 is set in the resource management table 56. For example, a job whose job execution ID is “1” (710) is executed, the CPU usage rate (average) of the resource amount is 30%, the number of instructions (average) is 800,000 / s, the memory processing is 64 Mbyte / s, the disk Reading is set to 10 MByte / s and disk writing is set to 10 MByte in the resource management table 56.

FIG. 22 is a diagram showing the flow of processing of the job execution agent unit 92.
In step S2201, the job execution agent unit 92 receives the job ID, job name, and command name corresponding to the job name of the job to be executed from the job execution unit 25, and proceeds to step S2202. In step S2202, the resource information acquisition agent unit 91 is notified of information on a command name for monitoring the resource usage (sets job information for acquiring the resource amount), and the process advances to step S2203.
In step S2203, the job having the command name accepted in step S2201 is executed, the execution result of the command is notified to the job execution unit 25, and the process ends.

FIG. 23 is a diagram showing a screen for displaying the result of the resource amount used by the service to the operation manager.
In FIG. 23, there are a resource management tab 2301, a configuration management tab 2302, a job management tab 2303, and an event management tab 2304. In the resource management tab 2301, the amount of resources used by the job is displayed in a display area 2305. The virtual machine name and service name for displaying the resource amount can be selected by the item selection unit 2306. Further, the resource items of the job displayed in the display area 2305 can be changed by the buttons 2308 to 2310.

  In addition, the job resource information displayed in the display area 2305 is output separately for each item of the job execution type 2313. The job information 2311 displays the execution ID and job name of the executed job. The configuration information 2312 displays the name of the host machine that executed the job, and the job execution type 2313 displays the job execution type. The event information 2314 displays event information that causes the job to be executed. As the total item 2315, an item to be totaled can be selected for display items (job information 2311 to event information 2314). For example, the resource usage amount of the job executed by the countermeasure 2 can be displayed by selecting the countermeasure 2 in the event information item. In addition, by specifying the host machine 1 in the configuration information 2312, it is possible to output a list of used resources of jobs executed on the host machine 1. In this way, the items that the operation manager wants to output can be selected centrally, and the resource usage can be output.

  As described above, according to the service / resource management system S of the present embodiment, even when the shared resource allocated to the virtual machine is dynamically changed or the virtual machine is moved to another business computer, the service / resource management system S Resource usage is managed, and when moving a virtual machine, business computers that satisfy the resource usage amount of the service, or the execution order of the service are controlled to satisfy the resource usage amount for each service before the end date The destination of the business computer can be determined. In addition, by managing the amount of resources used by the service, including the cause of a virtual machine configuration change, such as the occurrence of a failure, and the re-execution (rerun) of a job, the resource usage can be displayed for each factor.

  This is the end of the description of the embodiments, but the aspects of the present invention are not limited to these. For example, the function of the event management device 4 may be absorbed by the service management device 2. Specific configurations of hardware, programs, and the like can be changed as appropriate without departing from the spirit of the present invention.

It is a figure which shows the whole structure of the service resource management system of this embodiment. It is a figure which shows the hardware constitutions of the service management apparatus and service resource monitoring apparatus of this embodiment. It is a figure which shows the table stored in the memory | storage part of the service management apparatus of this embodiment. It is a figure which shows the hardware constitutions of the business computer which comprises the event management apparatus of this embodiment, and a business. It is a figure which shows the system configuration | structure of this embodiment. It is a figure which shows the structure of the business computer of this embodiment. It is a figure which shows the event handling table of this embodiment. It is a figure which shows the event management table of this embodiment. It is a figure which shows the service structure table of this embodiment. It is a figure which shows the job management table of this embodiment. It is a figure which shows the machine configuration table of this embodiment. It is a figure which shows the performance coefficient table of this embodiment. It is a figure which shows the virtual machine structure table of this embodiment. It is a figure which shows the resource management table of this embodiment. It is a figure which shows the job history management table of this embodiment. It is a flowchart which shows the flow of a process which acquires the resource structure of the business computer of this embodiment. It is a flowchart which shows the flow of the process which creates the virtual machine of this embodiment. It is a flowchart which shows the flow of the movement process of the virtual machine of this embodiment. It is a flowchart which shows the flow of the process which determines the movement destination host machine of the virtual machine of this embodiment. It is a flowchart which shows the flow of the process which determines the movement destination host machine of the virtual machine of this embodiment. It is a flowchart which shows the flow of the process which determines the movement destination host machine by the change of the execution order of the service on the virtual machine of this embodiment. It is a flowchart which shows the flow of a job execution process of this embodiment. It is a flowchart which shows the flow of a process of the job execution agent of this embodiment. It is a screen which displays the result of the resource amount used by the service of this embodiment.

Explanation of symbols

1 Service / Resource Monitoring Device 2 Service Management Device 3 Network 4 Event Management Device 5 Storage Unit 1
6 Storage unit 2
7 Business computer 8 Host machine 9 Virtual machine 10 GUI
DESCRIPTION OF SYMBOLS 11 Service management program 12 Event management program 13 Business system 20 Service configuration management part 21 Virtual machine generation change part 22 Job management part 23 Resource information acquisition part 24 Service configuration change part 25 Job execution part 26 Resource information display part 27 Configuration information display part 41 Event Management Unit 42 Event Analysis Unit 43 Event Handling Execution Unit 44 Event Display Unit 51 Service Configuration Table 52 Job Management Table 53 Machine Configuration Table 54 Performance Coefficient Table 55 Virtual Machine Configuration Table 56 Resource Management Table 57 Job History Management Table 61 Event Response Table 62 Event management table 81 Virtual machine generation change agent unit 82 Configuration acquisition agent unit 83 Configuration change agent unit 91 Resource information acquisition agent Gent part 92 Job execution agent part 93 Event notification agent part 94 Business program

Claims (10)

A plurality of virtual machines that provide services on one or more business computers and are movable on the one or more business computers, and a host that is deployed for each business computer and manages the virtual machines in its business computer A virtual system environment comprising a machine, managing a service configuration and shared resources of the business computer used by the service, determining the business computer that can be moved from the resource usage when the virtual machine is moved, A service management device for moving a virtual machine,
Service configuration information including information on the expiration date of the service, job management information including information on a job to be executed upon restart after each job that configures the service, and host machine configuration indicating the configuration of the host machine Information, virtual machine configuration information indicating the configuration of the virtual machine, resource management information including information on the executed job and the amount of resources used by the job, and the type of event that has occurred in the virtual machine A storage unit for storing event information including coping information;
When moving the virtual machine, the virtual machine operates on the virtual machine based on the service configuration information, job management information, host machine configuration information, virtual machine configuration information, resource management information, and event information stored in the storage unit. A processing unit that determines and moves the destination of the virtual machine so as to satisfy the condition that the existing service is completed by the end deadline;
A service management apparatus comprising: The storage unit
Job history management information including correspondence information between the job execution history and the event that causes the job to be executed is further stored,
The processor is
The resource usage amount for the execution of the job is totaled by referring to job history management information stored in the storage unit and totaling the resource usage amount for each event. Service management device. The processor is
Corresponding information between the change history of the configuration of the shared resource managed by the host machine and the configuration of the shared resource at the time of execution of the job due to the movement of the virtual machine is newly stored in the machine configuration information. The service management apparatus according to claim 1. The storage unit
Performance coefficient information indicating the performance relationship between the host machine and the shared resource managed by the host machine is further stored,
The processor is
The service management apparatus according to claim 1, wherein when determining the migration destination of the virtual machine, the migration destination of the virtual machine is determined so as to satisfy the condition by referring to the performance coefficient information. The processor is
After selecting a candidate for the migration destination of the virtual machine, determine whether the host computer of the destination business computer can secure the necessary resource performance. The service management apparatus according to claim 1, wherein the service management apparatus is moved. The processor is
When determining the migration destination of the virtual machine, if the execution order of services executed after the virtual machine is moved is changed, it is determined whether or not the processing of all services is completed earlier. The service management apparatus according to claim 1, wherein the execution order is changed. The processor is
The resource amount required by the moved virtual machine is allocated to the virtual machine from the shared resources managed by the host machine of the destination business computer after the virtual machine is moved. Service management equipment. A plurality of virtual machines that provide services on one or more business computers and are movable on the one or more business computers, and a host that is deployed for each business computer and manages the virtual machines in its business computer A virtual system environment comprising a machine,
A service management apparatus that manages a service configuration and shared resources of the business computer used by a service, determines the business computer that can be moved from the resource usage when the virtual machine is moved, and moves the virtual machine,
Service configuration information including information on the expiration date of the service, job management information including information on a job to be executed upon restart after each job that configures the service, and host machine configuration indicating the configuration of the host machine Information, virtual machine configuration information indicating the configuration of the virtual machine, resource management information including information on the executed job and the amount of resources used by the job, and the type of event that has occurred in the virtual machine A service management method for managing a service based on the storage content of a storage unit that stores event information including handling information,
The processing unit of the service management device includes:
When moving the virtual machine, the virtual machine operates on the virtual machine based on the service configuration information, job management information, host machine configuration information, virtual machine configuration information, resource management information, and event information stored in the storage unit. A service management method comprising: deciding a migration destination of a virtual machine and moving the virtual machine so as to satisfy a condition that a certain service is completed by the expiration date. The storage unit
Job history management information including correspondence information between the job execution history and the event that causes the job to be executed is further stored,
The processor is
The resource usage amount for the execution of the job is totaled by referring to job history management information stored in the storage unit and totaling the resource usage amount for each event. Service management method. One or more business computers,
Service configuration for a virtual system environment comprising a plurality of virtual machines that can be moved on the one or more business computers, and a host machine that is deployed for each business computer and manages the virtual machines in its business computers And a shared resource of the business computer used by the service, determining the business computer that can be moved from the resource usage when the virtual machine is moved, and a service management device that moves the virtual machine;
A service management system comprising:
The service management device includes:
Service configuration information including information on the expiration date of the service, job management information including information on a job to be executed upon restart after each job that configures the service, and host machine configuration indicating the configuration of the host machine Information, virtual machine configuration information indicating the configuration of the virtual machine, resource management information including information on the executed job and the amount of resources used by the job, and the type of event that has occurred in the virtual machine A storage unit for storing event information including coping information;
When moving the virtual machine, the virtual machine operates on the virtual machine based on the service configuration information, job management information, host machine configuration information, virtual machine configuration information, resource management information, and event information stored in the storage unit. A processing unit that determines and moves the destination of the virtual machine so as to satisfy the condition that the existing service is completed by the end deadline;
A service management system comprising:

Images