JP2011100263A - Virtual computer system, virtual computer management method and management program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a virtual computer system configured to easily change the structure of a job flow or data in a computer virtualized environment, to facilitate batch processing. <P>SOLUTION: The system includes: a virtual machine control means 151 which changes the configuration of a plurality of virtual computers; a job flow generation means 152 which changes, in response to a configuration change of virtual computers, the virtual computers to execute each job; a data management means 154 which assigns a data file corresponding to a data label name preliminarily defined as a data file to be formed or referred to in each job, to a shared storage; a storage means 102 which stores a data location database 170 storing location information of the data file in association with the data label name; and a job flow execution means 153 which refers to the data management means for the location information corresponding to the data label name, and transmits it to the virtual computer which executes each job. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to execution of a job flow on a computer, and more particularly to execution of a job flow on a computer in a virtual environment.

  A job flow refers to the flow of a plurality of jobs executed on a computer, with the execution order and association being associated with each other. One job flow may include parallel execution in which a plurality of jobs are executed at the same time, conditional branching in which subsequent execution contents change based on the execution result of a specific job, and the like. For example, this job flow process is used for such purposes as performing routine batch processing on the computer at night or on holidays when there are few users of the computer.

  In companies, computer virtualization environments are often used. If you can operate a virtual computer in the data center and perform processing by job flow, batch processing in various departments of the company can be concentrated in one data center and executed efficiently. It is thought that the time, cost, power consumption, etc. required for execution can be reduced. In addition, since data backup processing can be performed in a lump, it is considered that it is possible to easily cope with a failure or the like.

  For example, when the load applied to batch processing is small, a plurality of virtual computers are combined into a small number of physical computers, and the number of physical computers that are actually energized and operated is reduced. On the other hand, when a large load is applied to a specific virtual computer, the number of operating physical computers is increased and more virtual computers are operated to increase computer resources that can be used for the processing. In the computer virtualization environment, it is considered that the dynamic change of the configuration of such virtual computers can be easily performed, and thus the efficiency of execution can be improved.

  Technologies that may be related to this include: Patent Document 1 describes a linkage instruction information generation system in which when a device that can execute a job is not on the network, an alternative device can be searched and an instruction can be changed and executed if necessary. Yes. Japanese Patent Application Laid-Open No. 2004-228561 describes a creation device that can create and execute a job flow using a GUI (Graphic User Interface).

Japanese Patent Laying-Open No. 2005-077853 Japanese Patent Application Laid-Open No. 09-044346

  However, the job flow related to batch processing is created specifically for processing in each business, and the data used in the processing also has a configuration specialized for processing in each business. Therefore, when the configuration of the virtual computer that executes the batch processing is dynamically changed, it is necessary to change the job flow and the data configuration accordingly. It is difficult to automate this configuration change.

  In particular, the standardized batch process is required to be executed without manpower since it is required to be executed at night or on holidays when there are few users of the computer as described above. Accordingly, if there is a change in the job flow or data configuration that is difficult to automate due to such a change in the configuration of the virtual computer, it is difficult to perform the processing itself by batch processing. If this change cannot be made easily, it is difficult to efficiently execute this job flow in a virtual environment.

  None of the inventions described in Patent Documents 1 and 2 relate to a job flow and a data configuration change accompanying a configuration change of a virtual computer, and thus do not solve this problem.

  It is an object of the present invention to easily execute batch processing by easily changing the job flow and data configuration accompanying the dynamic configuration of the virtual computer in a computer virtualization environment. A virtual computer system, a virtual computer management method, and a management program.

  To achieve the above object, a virtual computer management device according to the present invention includes a physical machine that is a physical computer device capable of operating a virtual computer and a common storage that is a storage device accessible from the virtual computer, A management server for managing the operation of the virtual computer on the physical machine, the management server instructing to set and change a plurality of virtual computers operating on the physical machine; Job flow generation means for changing a virtual computer that executes each job in a job flow configured in advance by a plurality of jobs that can be executed by the virtual computer in response to receiving a setting and change command of the computer, and Data files created or referenced by other jobs Data management means for allocating a data file corresponding to a data label name defined in advance on the common storage, and storage means for storing a data storage location database for storing data file storage location information in association with the data label name And job flow execution means for inquiring the storage location information corresponding to the data label name to the data management means and transmitting the inquiry to the virtual computer that executes the job.

  To achieve the above object, a virtual computer management method according to the present invention includes a physical machine that is a physical computer device capable of operating a virtual computer, and a common storage that is a storage device accessible from the virtual computer. In a virtual computer system interconnected with a management server that manages the operation of a virtual computer on a physical machine, the virtual machine control means instructs setting and changing of a plurality of virtual computers operating on the physical machine. The job flow generation means changes the virtual computer that executes each job in a job flow that is configured in advance by a plurality of jobs that can be executed by the virtual computer in response to the configuration change of the virtual computer, and is created by each job. Or pre-defined as referenced data file The data management means allocates the data file corresponding to the specified data label name on the common storage, stores the storage location information of the data file in association with the data label name, and stores the storage location information corresponding to the data label name in the job flow. The execution means transmits the job to a virtual computer that executes the job.

  To achieve the above object, a virtual computer management program according to the present invention includes a physical machine that is a physical computer device capable of operating a virtual computer, and a common storage that is a storage device accessible from the virtual computer. In a virtual computer system interconnected with a management server that manages the operation of virtual computers on a physical machine, the configuration of multiple virtual computers that operate on the physical machine is changed to a computer that the management server has A procedure for changing a virtual computer that executes each job in a job flow pre-configured by a plurality of jobs that can be executed by the virtual computer in response to a change in the configuration of the virtual computer, created or referenced in each job Predefined as a data file Execute the job to assign the data file corresponding to the specified data label name on the common storage, to store the data file storage location information corresponding to the data label name, and the storage location information corresponding to the data label name A procedure for transmitting to a virtual computer is executed.

  As described above, according to the present invention, since a data label name is assigned to a data file created by each job constituting the job flow, and thus the virtual computer accesses the data file, each job is configured. Even if the configuration of the virtual computer to be executed is changed, it is possible to easily change the job flow and data configuration associated therewith, and this configuration change can be automated. Accordingly, it is possible to provide a virtual computer system, a virtual computer management method, and a management program having an excellent feature that batch processing can be easily executed.

It is explanatory drawing which shows the structure of the data center which concerns on the 1st Embodiment of this invention. FIG. 2 is an explanatory diagram showing more detailed configurations of a job flow generation unit, a job flow execution unit, a data management unit, a job information database, and a data storage location database shown in FIG. 1. FIG. 3 is an explanatory diagram showing a data configuration of a server usage information table shown in FIG. 2 and a specific example thereof. It is a flowchart which shows the detailed operation | movement of the server use confirmation part shown in FIG. FIG. 3 is an explanatory diagram showing a data configuration of a use-specific execution job information table shown in FIG. 2 and a specific example thereof. 3 is a flowchart showing a detailed operation of an execution job extraction unit shown in FIG. 2. FIG. 3 is an explanatory diagram illustrating a data configuration of the job parallelism information table illustrated in FIG. 2 and a specific example thereof. 3 is a flowchart showing a detailed operation of a job parallelism confirmation unit shown in FIG. 2. FIG. 3 is an explanatory diagram illustrating a data configuration of a job dependency relationship information table and a job execution information table illustrated in FIG. 2 and a specific example thereof. 3 is a flowchart showing a detailed operation of a job dependency confirmation unit shown in FIG. 2. FIG. 3 is an explanatory diagram illustrating a specific example of job flow information illustrated in FIG. 2. 11A shows job flow information before the series of processing shown in FIGS. 4, 6, 8, and 10 is performed, and FIG. 11B shows job flow information after the processing is performed. . 3 is a flowchart showing a detailed operation of a job flow control unit shown in FIG. 2. 3 is a flowchart showing detailed operations of a job input unit shown in FIG. 2. It is explanatory drawing which shows the more detailed structure of the data storage location database shown to FIGS. It is a flowchart which shows the detailed operation | movement of the data storage location allocation part shown in FIG. 3 is a flowchart showing detailed operations of a data storage location inquiry unit shown in FIG. 2. It is explanatory drawing which shows the structure of the data center which concerns on the 2nd Embodiment of this invention. It is explanatory drawing which shows the more detailed structure of the job flow production | generation means shown in FIG. 17, a job flow execution means, a data management means, a job information database, and a data storage location database. It is explanatory drawing which shows the data structure of the data storage location table shown in FIG. 18, and a label reference information table, and its specific example. It is a flowchart which shows the detailed operation | movement of the data label count part shown in FIG. It is a flowchart which shows the detailed operation | movement of the data storage location allocation part shown in FIG. It is a flowchart which shows the detailed operation | movement of the data storage location inquiry part shown in FIG.

(First embodiment)
Hereinafter, the structure of the 1st Embodiment of this invention is demonstrated based on attached FIGS. 1-2.
First, the basic content of the present embodiment will be described, and then more specific content will be described.
The virtual computer system (data center 10) according to the present embodiment includes physical machines 111 to 11m, which are physical computer devices capable of operating a virtual computer, and a common storage 131, which is a storage device accessible from the virtual computer. And a management server 100 that manages the operation of the virtual computer on the physical machine. The management server 100 corresponds to a virtual machine control unit 151 that commands setting and changing of a plurality of virtual computers operating on a physical machine, and a command to set and change a virtual computer from the virtual machine control unit. A job flow generation unit 152 that changes a virtual computer that executes each job in a job flow that is configured in advance by a plurality of jobs that can be executed by the virtual computer, and a data file that is created or referred to by each job in advance Data management means 154 for allocating a data file corresponding to the defined data label name on the shared storage, and storage means 102 for storing a data storage location database 170 for storing data file storage location information in association with the data label name And the data label name And a job flow execution unit 153 to transmit the location information to respond to a virtual machine that executes the job so as to query the data management means.

  The storage unit 102 stores a server usage information table 161 for storing the usage of the virtual computer, an execution job corresponding to the usage of the virtual computer, and a data label name of a data file created or referred to by the execution job. A server usage confirmation unit that stores a job information database 160 including an execution job information table 162 and in which the job flow generation unit 152 refers to the server usage information table and confirms the usage of the virtual computer whose configuration has been changed by the virtual machine control unit. 152a, an execution job extraction unit 152b that extracts a data label name corresponding to the usage of the virtual computer whose configuration has been changed by referring to the usage-specific execution job information table, and job flow information corresponding to the virtual computer whose configuration has been changed. Change the job And a dependency check unit 152d.

The job information database 160 further includes a job parallelism information table 163 that stores whether or not each job constituting the job flow can be executed in parallel.
The job flow generation unit 152 includes a job dependency confirmation unit 152c that refers to the job parallel degree information table 163 and changes job flow information so that jobs that can be executed in parallel are executed in parallel.

With the above configuration, the virtual computer management apparatus 100 can easily change the job flow and data configuration associated therewith even when the configuration of the virtual computer is changed.
Hereinafter, this will be described in more detail.

  FIG. 1 is an explanatory diagram showing the configuration of the data center 10 according to the first embodiment of the present invention. The data center 10 includes a management server 100 that manages the entire data center 10, m physical machines 111 to 11 m that are physical computer devices on which virtual machines to be described later are constructed by management of the management server 100, and a management server 100 and a file server 130 accessible from the physical machines 111 to 11m are connected to each other by an internal network 140.

  On the physical machines 111 to 11m, n virtual machines 121 to 12n are constructed by management of the management server 100. The file server 130 can be accessed from any of the virtual machines 121 to 12n. Here, m and n are both integers, and m = n or m ≠ n. Further, the number of virtual machines constructed for each physical machine may be different.

  The management server 100 is a computer device that includes a processor 101 that is an arithmetic device that executes a computer program, and a storage device 102 that stores programs and data. In the processor 101, each unit such as a virtual machine control unit 151, a job flow generation unit 152, a job flow execution unit 153, and a data management unit 154, which will be described later, operates as a computer program.

  Further, the storage device 102 stores a job information database 160 used by the job flow generation unit 152 and job flow execution unit 153 and a data storage location database 170 used by the data management unit 154. The detailed contents of each database will also be described later.

  The virtual machine control unit 151 performs an operation of adding and deleting the virtual machines 121 to 12n on the physical machines 111 to 11m. This can be performed dynamically according to the load of the virtual machines 121 to 12n. In some cases, the physical machines 111 to 11m may be started and stopped. When the virtual machines 121 to 12n are added or deleted, the virtual machine control unit 151 notifies the job flow generation unit 152 to that effect.

  The job flow generation unit 152 generates and changes the job flow in response to the notification of addition or deletion of the virtual machines 121 to 12n from the virtual machine control unit 151, and passes the job flow to the job flow execution unit 153. Let it run. Then, when executing the job flow, the job flow execution means 153 inquires the data management means 154 about the location of the data file to be referred to or created, and refers to or creates the data file at the position returned in response to the inquiry. Do.

  The file server 130 includes a common storage 131 that is a storage device accessible from all of the management server 100, the physical machines 111 to 11m, and the virtual machines 121 to 12n. In the present embodiment, an example is shown in which an actual data file is stored in the common storage 131, but the data file is located on a storage device accessible by the management server 100 and the virtual machines 121 to 12n. Anywhere.

  FIG. 2 is an explanatory diagram showing a more detailed configuration of the job flow generation unit 152, job flow execution unit 153, data management unit 154, job information database 160, and data storage location database 170 shown in FIG. The job flow generation unit 152 includes functional units such as a server usage confirmation unit 152a, an execution job extraction unit 152b, a job parallelism confirmation unit 152c, and a job dependency confirmation unit 152d.

  The job flow execution unit 153 includes functional units such as a job flow control unit 153a and a job input unit 153b. The data management unit 154 includes functional units such as a data storage location assignment unit 154a and a data storage location inquiry unit 154b. The operations of these functional units will be described later.

  The job information database 160 includes each table and information such as a server usage information table 161, a usage-specific execution job information table 162, a job parallelism information table 163, a job dependency relationship information table 164, a job execution information table 165, and job flow information 166. Including. The data storage location database 170 includes a data storage location table 171. The data stored in each of these and specific examples thereof will also be described later.

(Job flow generation means)
FIG. 3 is an explanatory diagram showing a data configuration of the server usage information table 161 shown in FIG. 2 and a specific example thereof. The server usage information table 161 stores a server name 161a allocated to each of the virtual machines 121 to 12n, and a usage 161b assigned to each of the server names 161a. The virtual machine control unit 151 adds the server name 161a and the usage 161b of the added server to the server usage information table 161 in the case of the addition notification in response to the addition notification or deletion notification of the virtual machines 121 to 12n. In the case of a deletion notification, the server name 161a and its usage 161b are deleted for the deleted server.

  FIG. 4 is a flowchart showing a detailed operation of the server usage confirmation unit 152a shown in FIG. The server usage confirmation unit 152a reads the server usage information table 161 (step S201), determines the usage 161b corresponding to each of the virtual machines 121 to 12n (step S202), and passes the processing to the execution job extraction unit 152b.

  FIG. 5 is an explanatory diagram showing a data configuration of the execution job information table for each use 162 shown in FIG. 2 and a specific example thereof. The execution job information table 162 classified by usage includes the usage 161b of each server recorded in the server usage information table 161, an execution batch job 162a corresponding to the usage 161b, and a data file to be created when the execution batch job 162a is executed. Creation data information 162b indicating the data label name and reference data information 162c indicating the data label name of the data file to be referred to are stored.

  The “data label name” here is an arbitrary name that can be uniquely indicated for the data file, and does not need to match the file name or directory name of the actual data file.

  FIG. 6 is a flowchart showing a detailed operation of the execution job extraction unit 152b shown in FIG. The execution job extraction unit 152b reads the application-specific execution job information table 162 (step S211), and executes the execution batch job 162a to be executed by each server and the creation data information 162b which is the data label name created and referred to at that time. The reference data information 162c is extracted (step S212), and the process is passed to the job parallelism confirmation unit 152c.

  FIG. 7 is an explanatory diagram showing a data configuration of the job parallelism information table 163 shown in FIG. 2 and a specific example thereof. The job parallelism information table 163 stores an execution batch job 162a recorded in the application-specific execution job information table 162 and a parallel execution availability 163a indicating whether or not the job can be executed in parallel.

  FIG. 8 is a flowchart showing a detailed operation of the job parallelism confirmation unit 152c shown in FIG. The job parallelism confirmation unit 152c first reads the job parallelism information table 163 (step S221), and determines whether there are a plurality of jobs to be executed on the same server for each server (step S222). If not, the process ends. If there is a plurality, it is determined whether or not the parallel execution permission / inhibition 163a indicates that the parallel execution is possible (step S223).

  If parallel execution is “possible” in step S223, the job flow information 166 is changed to execute in parallel (step S224), and the process is terminated. If the parallel execution is not possible, the process ends without changing the job flow information 166. After the above processing is completed, the processing is passed to the job dependency confirmation unit 152d.

  FIG. 9 is an explanatory diagram showing the data structure of the job dependency relationship information table 164 and the job execution information table 165 shown in FIG. 2 and specific examples thereof. The job dependency relationship information table 164 stores an execution batch job 162a recorded in the execution job information table 162 for each use and a preceding batch job 164a that must be performed prior to the job. In the job execution information table 165, execution batch job 162a recorded in the use job execution information table 162, that is, executed information 165a indicating whether each job constituting the job flow has been executed is recorded. ing.

  FIG. 10 is a flowchart showing a detailed operation of the job dependency confirmation unit 152d shown in FIG. The job dependency confirmation unit 152d first reads the job dependency information table 164 and the job execution information table 165 (step S231), and the notification received from the virtual machine control unit 151 is an additional notification of the virtual machines 121 to 12n. It is determined whether the notification is a deletion notification (step S232). If it is a deletion notification, job flow information 166 is created by deleting the execution batch job 162a corresponding to the deleted use 161b of the virtual machine (step S233), and the process is terminated.

  If it is an addition notification, the job dependency confirmation unit 152d determines that the preceding batch job 164a corresponding to the execution batch job 162a corresponding to the added usage 161b of the virtual machine is “-(none)” or this preceding batch job 164. It is determined whether or not the executed information 165a corresponding to the above corresponds to any of “executed” (step S234).

  If the preceding batch job 164a is “-(none)” or the corresponding executed information 165a is “executed”, job flow information 166 to which the executed batch job 162a is added is created (step S235). The process ends. If the preceding batch job 164a exists and the executed information 165a corresponding to the preceding batch job 164a is “not executed”, the process ends without creating or changing the job flow information 166.

  FIG. 11 is an explanatory diagram showing a specific example of the job flow information 166 shown in FIG. FIG. 11A shows job flow information 166 before the series of processing shown in FIGS. The job flow information 166 here is that the usage 161b is “AP (application) server”, the execution batch job 162a is “form data creation”, the server name 161a “server A”, and the usage 161b is “sorting server”. The processing is transferred to the server name 161a “server C” whose batch job 162a is “data sorting”, and thereafter the server name 161a “server D” whose usage 161b is “form server” and whose execution batch job 162a is “form creation”. A series of operations is shown in which processing is passed to and execution is terminated.

  Data corresponding to this example is stored in each table shown in FIG. 3, FIG. 5, FIG. 7, and FIG. In this state, the job flow generation unit 152 receives a “virtual machine addition notification” from the virtual machine control unit 151. More specifically, the notification is that a virtual machine “server B” whose usage is “AP server” is added. The virtual machine control unit 151 adds “server B” to the server name 161 a corresponding to “AP server” in the usage 161 b of the server usage information table 161 together with this notification.

  As shown in FIG. 4, when the server usage confirmation unit 152a reads the server usage information table 161 (step S201), the usage 161b of the newly added server name 161a “Server B” is “AP server” ( Step S202). Upon receiving this information, the execution job extraction unit 152b reads the application-specific execution job information table 162 as shown in FIG. 6 (step S211), and executes the execution batch job 162a “form” corresponding to the application 161b “AP server”. “Create data”, created data information 162b “data file A”, and reference data information 162c “-(none)” are extracted (step S212).

  Upon receiving this information, the job parallelism confirmation unit 152c reads the job parallelism information table 163 as shown in FIG. 8 (step S221), and determines whether or not there are a plurality of jobs to be executed in each server. Judgment is made (step S222). In this example, since there are no jobs other than “create form data” other than “server A” and “server B”, which are “AP servers”, the process is terminated and the process is transferred to the job dependency confirmation unit 152d.

  As shown in FIG. 10, the job dependency confirmation unit 152d to which the process has been passed reads the job dependency information table 164 and the job execution information table 165 (step S231), and receives the notification received from the virtual machine control unit 151. Is an addition notification of the virtual machine “server B” (step S232), and the preceding batch job 164 corresponding to the “execution batch job 162a“ form data creation ”corresponding to the usage 161b“ AP server ”of the“ server B ”is“ -(None) "(step S234), the job flow information 166 to which this" server B "is added is created (step S235), and the process is terminated.

  FIG. 11B shows job flow information 166 after the processing shown in FIGS. This is because “server B”, which is “AP server” for performing “form data creation”, is added to the job flow information 166 shown in FIG. 11A by performing the processing shown in FIG. This is the job flow information 166 in the state after the check.

(Job flow execution means)
FIG. 12 is a flowchart showing detailed operations of the job flow control unit 153a shown in FIG. First, the job flow control unit 153a reads the job flow information 166, and also reads the executed information 165a of the job execution information table 165 (step S251). Is determined as an execution job (step S252).

  The job flow control unit 153a passes this execution job to the job input unit 153b and causes the corresponding server name 161a to execute the job (step S253), and waits for the execution job to be completed (step S254). Then, it is determined whether or not the job is the final job in the order described in the job flow information 166 (step S255). If the job is the final job, execution of the job flow is ended. If it is not the final job, the process returns to step S252 to execute the subsequent job.

  FIG. 13 is a flowchart showing detailed operations of the job input unit 153b shown in FIG. The job input unit 153b determines whether or not the execution job (execution batch job 162a) passed from the job flow control unit 153a is a job accompanied by data creation, that is, the creation data information 162b in the use-specific execution job information table 162. Is a registered job (step S261). If the created data information 162b is registered, the data storage location allocation unit 154a is requested to allocate the storage location of the data (step S262).

  Similarly, the job input unit 153b determines whether or not this execution job is a job with reference to data, that is, whether or not the reference data information 162c is registered in the execution job information table 162 for each use. Is determined (step S263). If the reference data information 162c is registered, the data storage location inquiry unit 154b is inquired about where the reference data is registered (step S264).

  Since the data storage location allocation unit 154a and the data storage location inquiry unit 154b return information about the data storage location, the job input unit 153b sends the execution job to the server of the corresponding server name 161a based on this information. This is executed (step S265).

(Data management means)
FIG. 14 is an explanatory diagram showing a more detailed configuration of the data storage location database 170 shown in FIGS. The data storage location database 170 includes a data storage location table 171. The data storage location table 171 is a storage location indicating the data label name 171a corresponding to the created data information 162b or the reference data information 162c and the location (computer name, directory name, file name, etc.) where the data is actually stored. Information 171b.

  FIG. 15 is a flowchart showing a detailed operation of the data storage location assignment unit 154a shown in FIG. When the data storage location allocation unit 154a receives a data storage location allocation request from the job input unit 153b in step S262 (step S271), the data storage location allocation unit 154a allocates the data storage location as the data label name 171a. The data is stored in the data storage location table 171 (step S272), and the storage location information 171b is returned to the job input unit 153b (step S273).

  FIG. 16 is a flowchart showing a detailed operation of the data storage location inquiry unit 154b shown in FIG. When the data storage location inquiry unit 154b receives an inquiry about the data storage location from the job input unit 153b in step S264 (step S281), the data storage location inquiry unit 154b searches the received data reference name 162c from the data label name 171a. The storage location information 171b corresponding to the data label name 171a hit in the search is returned to the job input unit 153b (step S282).

  In the above example, even when a virtual machine called “Server B”, which is an “AP server” in charge of “form data creation”, is added, the execution batch job 162 a “form data creation” has the created data information 162 b. Since “data A” is to be created, the job flow control unit 153a that executes this job flow reads the job flow information 166 of this job flow (step S251), and “form data creation” described at the beginning of the job flow is performed. "Is executed (step S252), and the execution job is transferred to the job input unit 153b (step S253).

  The job input unit 153b that has received the “form data creation” job is a job in which “data A” is registered in the created data information 162b (step S261), and therefore this data is stored in the data storage location allocation unit 154a. A request is made to allocate a storage location for “data A” (step S262). Upon receiving this request, the data storage location assignment unit 154a assigns the data storage location corresponding to the data label 171a as “data A”, and sends the actually assigned storage location information 171b to the job input unit 153b. Return (step S272).

  In response to this, the job input unit 153b passes the storage location information 171b to the corresponding server names 161a “server A” and “server B”, and executes the job of the execution batch job 162a “form data creation”. (Step S265). If this can be executed, the job flow control unit 153a and the job input unit 153b sequentially execute the subsequent “data sorting” and “form creation” jobs.

  As described in the above example, even when a virtual machine “server B” is added, there is no change in the job dependency before and after the addition, and there is no change in the data reference relationship. However, only by adding “Server B” to the server name 161a corresponding to the application 161b “AP server” to be newly assigned in the server application information table 161, the data in the other tables is changed. No.

  That is, only by this one change, a “form data creation” job handled by the “AP server” defined in the job flow information 166 can be performed only by “server A” as shown in FIG. From execution, as shown in FIG. 11 (b), it can be changed without any problem so that it is executed in parallel on two servers "server A" and "server B". Conversely, “server B” is deleted from the state of the job flow information 166 shown in FIG. 11B, and the “form data creation” job is executed only by “server A” as shown in FIG. 11A. The same operation is possible for the case of making them.

(Overall operation of the first embodiment)
Next, the overall operation of the above embodiment will be described. The virtual computer management method according to the present embodiment includes physical machines 111 to 11m that are physical computer devices capable of operating a virtual computer, a shared storage 131 that is a storage device accessible from the virtual computer, In a virtual computer system interconnected with a management server 100 that manages the operation of a virtual computer on a machine, the virtual machine control means 151 instructs setting and changing of a plurality of virtual computers operating on a physical machine. Then, the job flow generation unit 152 changes the virtual computer that executes each job in a job flow configured in advance by a plurality of jobs that can be executed by the virtual computer in response to the change in the configuration of the virtual computer. Data files to be created or referenced The data management unit 154 allocates the data file corresponding to the data label name previously defined on the shared storage, stores the data file storage location information in association with the data label name, and stores the data file in correspondence with the data label name. The job flow execution means 153 transmits the information to the virtual computer that executes the job.

Here, each of the above-described operation steps may be programmed to be executable by a computer, and may be executed by the management server 100 which is a computer that directly executes each of the steps.
With this configuration and operation, the present embodiment has the following effects.

  In the prior art, it has been necessary to make complicated changes in job dependency, data reference relationship, and the like as virtual machines are added or deleted. In the present embodiment, such complicated change processing is not required. Therefore, this embodiment can provide an environment suitable for batch processing that is required to be executed in a lump without manpower at night or on holidays. Since addition and deletion of virtual machines can be automatically performed based on the processing load and power consumption applied to each physical machine, it is possible to easily realize the efficiency by virtualization as described above.

(Second Embodiment)
In the virtual computer system (data center 310) according to the second embodiment of the present invention, in addition to the configuration of the first embodiment described above, the data storage location database 470 includes a job name, a virtual computer name, and data. The child number corresponding to the label name is stored together, and when the job flow generation unit 452 uses the same data label name for different jobs and virtual computers, the child number is assigned to the data storage location database. A data label count unit 452d to be stored is included.

  As a result, even when different jobs use data files with the same data label name, it is possible to prevent inconvenience associated with duplication of the data label name. This will be described in more detail below.

  FIG. 17 is an explanatory diagram showing the configuration of the data center 310 according to the second embodiment of this invention. Although the data center 310 has substantially the same configuration as the data center 10 according to the first embodiment described above, the operation of the management server that manages this data center is different, and this is called the management server 400. The physical machines 111 to 11m, the n virtual machines 121 to 12n, the file server 130, and the internal network 140 are the same as those in the first embodiment.

  The configuration of the management server 400 is completely the same as the management server 100 according to the first embodiment described above in terms of hardware and generally the same in terms of software, but job flow generation means that operates on the processor 101, The contents of the data management means and the data storage location database stored in the storage device 102 are different. These are referred to as job flow generation means 452, data management means 454, and data storage location database 470, respectively. Other virtual machine control means 151, job flow execution means 153, and job information database 160 are the same as those in the first embodiment.

  18 is an explanatory diagram showing more detailed configurations of the job flow generation unit 452, the job flow execution unit 153, the data management unit 454, the job information database 160, and the data storage location database 470 shown in FIG. The job flow generation unit 452 is in addition to the server usage confirmation unit 152a, execution job extraction unit 152b, job parallelism confirmation unit 152c, and job dependency confirmation unit 152d described in the first embodiment (these are the first implementations). Since the same operation as described in the embodiment is performed, the reference numbers are also completely the same), and there is a functional unit called a data label count unit 452e.

  The data management means 454 includes a data storage location assignment unit 454a and a data storage location inquiry unit 454b. The data storage location database 470 includes a data storage location table 471 and a label reference information table 472.

  FIG. 19 is an explanatory diagram showing the data structures of the data storage location table 471 and the label reference information table 472 shown in FIG. 18 and specific examples thereof. In the data storage location table 471, a child number 471a described later is recorded in addition to the data label name 171a and the storage location information 171b described in the first embodiment.

  The label reference information table 472 includes an execution batch job 162a recorded in the application-specific execution job information table 162, a server name 161a that executes the batch job, and the data label name 171a and child number 471a corresponding to the server name 161a. Is recorded.

  FIG. 20 is a flowchart showing a detailed operation of the data label count unit 452e shown in FIG. The data label counting unit 452e responds to the job dependency confirmation unit 152d creating the job flow information 166 by the operation of step S235 in FIG. 10, and the data corresponding to the added execution batch job 162a and the server name 161a. The label name 171a is stored in the label reference information table 472.

  At that time, first, it is determined whether or not the data of the same data label name 171a already exists in the label reference information table 472 (step S501). If it does not exist, the execution batch job 162a, server name 161a, and data label name are left as they are. 171a is stored in the label reference information table 472 (step S502). At this time, the child number 471a is set to an initial value such as “01”, for example.

  If data with the same data label name 171a already exists, it is determined whether the existing data, the execution batch job 162a, and the server name 161a are different (step S503). If they are different, the child number 471a is attached so as not to cover the existing data (for example, the child number 471a of the existing data is incremented by 1) (step S504), and the process proceeds to step S502 to store the data. If the execution batch job 162a and the server name 161a are not different with the same data label name 171a, the processing is ended as it is.

  FIG. 21 is a flowchart showing a detailed operation of the data storage location allocation unit 454a shown in FIG. When receiving a data storage location allocation request from the job input unit 153b in step S262 (step S271), the data storage location allocation unit 454a notifies the received created data information 162b as a data label name 171a at the same time. From the executed batch job 162a and the server name 161a, the label reference information table 472 is referred to and the corresponding child number 471a is searched (step S511).

  The data storage location is assigned corresponding to the data label name 171a and the child number 471a, the data is stored in the data storage location table 171 (step S512), and the storage location information 171b is returned to the job input unit 153b ( Step S273).

  FIG. 22 is a flowchart showing a detailed operation of the data storage location inquiry unit 454b shown in FIG. When the data storage location inquiry unit 154b receives an inquiry about the data storage location from the job input unit 153b in step S264 (step S281), the received reference data information 162c is notified as the data label name 171a at the same time. From the execution batch job 162a and the server name 161a, the label reference information table 472 is referred to and the corresponding child number 471a is searched (step S521).

  The storage location information 171b corresponding to the data label name 171a and the child number 471a is returned to the job input unit 153b (step S522). 21 to 22, the same processes as those illustrated in FIGS. 15 to 16 are denoted by the same reference numerals.

  In the present embodiment, even if different jobs are specified to create or refer to data files defined by the same data label name 171a, if the execution batch job 162a and the server name 161a are different, different child numbers are used. 471a is assigned and handled as separate data.

  In the data example shown in FIG. 19, an execution batch job 162a “form data creation” executed by the server name 161a “server A” and an execution batch job 162a “other” executed by the server name 161a “server X”. The data label name 171a “data A” was assigned to both. In this case as well, a separate data file is created with the child number 471a = “01” for the former and the child number 471a = “02” for the latter, so the data label name 171a is incurred. The resulting mistakes do not occur.

  The present invention has been described with reference to the specific embodiments shown in the drawings. However, the present invention is not limited to the embodiments shown in the drawings, and any known hitherto provided that the effects of the present invention are achieved. Even if it is a structure, it is employable.

  The present invention can be applied when a job flow is executed in a virtual environment of a computer.

10, 310 Data center 100, 400 Management server 101 Processor 102 Storage device 111, 11m Physical machine 121, 11n Virtual machine 130 File server 131 Common storage 140 Internal network 151 Virtual machine control means 152, 452 Job flow generation means 152a Server usage confirmation Unit 152b execution job extraction unit 152c job parallelism confirmation unit 152d job dependency confirmation unit 153 job flow execution unit 153a job flow control unit 153b job input unit 154, 454 data management unit 154a, 454a data storage location allocation unit 154b, 454b data Storage location inquiry section 160 Job information database 161 Server usage information table 161a Server name 161b Usage 162 Actual by usage Job information table 162a Execution batch job 162b Created data information 162c Reference data information 163 Job parallelism information table 163a Parallel execution availability 164 Job dependency information table 164a Predecessor batch job 165 Job execution information table 165a Executed information 166 Job flow information 170, 470 Data storage location database 171, 471 Data storage location table 171a Data label name 171b Storage location information 452e Data label count unit 471a Child number 472 Label reference information table

Claims (6)

A physical machine that is a physical computer device capable of operating a virtual computer, a common storage that is a storage device accessible from the virtual computer, and a management server that manages the operation of the virtual computer on the physical machine Prepared,
The management server is
Virtual machine control means for instructing setting and changing of the plurality of virtual computers operating on the physical machine;
The virtual computer that executes each job in a job flow that is configured in advance by a plurality of jobs that can be executed by the virtual computer in response to receiving a command to set and change the virtual computer from the virtual machine control unit Job flow generation means for changing
Data management means for allocating the data file corresponding to a data label name defined in advance as a data file created or referred to in each job on the common storage;
Storage means for storing a data storage location database for storing storage location information of the data file in association with the data label name;
A virtual computer system comprising: job flow execution means for querying the data management means for the storage location information corresponding to the data label name and transmitting the information to the virtual computer for executing the job. The storage means stores a server usage information table for storing the usage of the virtual computer, an execution job corresponding to the usage of the virtual computer, and the data label name of a data file created or referred to by the execution job. Stores job information database including job information table by usage
The job flow generation means refers to the server usage information table, refers to the server usage information table, and checks the usage of the virtual computer whose configuration has been changed by the virtual machine control means. An execution job extraction unit that extracts the data label name corresponding to the usage of the reconfigured virtual computer, and a job dependency confirmation unit that changes job flow information to correspond to the reconfigured virtual computer; The virtual computer system according to claim 1, comprising: The job information database includes a job parallelism information table for storing whether or not each job constituting the job flow can be executed in parallel;
The job flow generation unit includes a job dependency confirmation unit that changes the job flow information so that jobs that can be executed in parallel with reference to the job parallelism information table are executed in parallel. Item 3. The virtual computer system according to Item 2. The data storage location database stores both the job name, the virtual computer name, and the child number corresponding to the data label name,
A data label counting unit that, when the different job and the virtual computer use the same data label name, assigns a different child number and stores it in the data storage location database; The virtual computer system according to claim 1, further comprising: A physical machine that is a physical computer device capable of operating a virtual computer, a common storage that is a storage device accessible from the virtual computer, and a management server that manages the operation of the virtual computer on the physical machine In a virtual computer system connected to each other,
The virtual machine control means commands the setting and change of the plurality of virtual computers operating on the physical machine,
Job flow generation means changes the virtual computer that executes each job in a job flow configured in advance by a plurality of jobs that can be executed by the virtual computer in response to the configuration change of the virtual computer,
The data management unit allocates the data file corresponding to the data label name defined in advance as a data file created or referred to in each job on the common storage, and stores the storage location information of the data file in the data label name Remember to correspond to
A virtual computer management method, wherein job flow execution means transmits the storage location information corresponding to the data label name to the virtual computer that executes the job. A physical machine that is a physical computer device capable of operating a virtual computer, a common storage that is a storage device accessible from the virtual computer, and a management server that manages the operation of the virtual computer on the physical machine In a virtual computer system connected to each other,
In the computer provided in the management server,
A procedure for changing the configuration of a plurality of the virtual computers operating on the physical machine;
A procedure for changing the virtual computer that executes each job in a job flow configured in advance by a plurality of jobs that can be executed by the virtual computer in response to the configuration change of the virtual computer;
A procedure for allocating the data file corresponding to a data label name defined in advance as a data file created or referenced in each job on the common storage;
A procedure for storing storage location information of the data file in association with the data label name;
And a procedure for transmitting the storage location information corresponding to the data label name to the virtual computer that executes the job.

Images