JP2014099036A - Information processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve a problem associated with a conventional dynamic resource allocation method in which the priority of a virtual machine is determined in accordance with the priority of an application executed in the virtual machine to allocate the virtual machine dynamically on the basis of the priority, resulting the resource allocation after execution of a task and shortage of resources, causing delay in processing, if a task using a lot of resources is tried to execute suddenly.SOLUTION: When a user management part 121 detects a user 14 connecting to a virtual machine 13, a resource control part 122 allocates, before execution of a task, a resource to the virtual machine 13 to which the user 14 has connected, on the basis of resource allocation information managed at a resource management part 123.

Description

  The present invention relates to an information processing apparatus that efficiently allocates resources to a plurality of virtual machines.

  A virtualization technology for constructing a plurality of virtual servers by virtualizing one physical server is known. In such a virtualization technology, a virtual server is configured by constructing a virtual machine from hardware via virtualization software. For example, VMware ESXi (registered trademark) or Xen (registered trademark) is used as the virtualization software. )and so on.

  A plurality of constructed virtual machines can share and hold the resources of one physical server, and the resources of each virtual machine are designated and assigned. In such resource allocation, a method of dynamically allocating resources to each virtual machine is known in order to effectively use limited resources existing on a physical server (see, for example, Patent Document 1). ).

JP 2009-181578 A

  However, in the conventional dynamic resource allocation method, the priority of the virtual machine is determined according to the priority of the application running on the virtual machine, and the resource is dynamically allocated to the virtual machine based on the priority. However, resource allocation is after task execution, and even if an attempt is made to execute a task that uses a lot of resources suddenly, there is a problem that processing is delayed due to insufficient resources.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide an information processing apparatus and an information processing system capable of allocating resources before task execution and preventing processing delay due to resource shortage.

  An information processing apparatus according to the present invention is an information processing apparatus that controls resource allocation to a plurality of virtual machines, detects whether a user is connected to the virtual machine, and manages information on the connected user A user information management unit, a resource management unit that manages resources allocated to the plurality of virtual machines and surplus resources of the hardware as resource allocation information among hardware resources, and resources to the plurality of virtual machines A resource control unit to be allocated, and when the user information management unit detects a connection to the first virtual machine by the user, the resource control unit detects the first virtual machine based on information of the connected user. It is characterized by allocating resources to machines.

  According to the information processing apparatus of the present invention, the connection to the virtual machine of the user is detected, and resources are allocated to the connected virtual machine before executing the task. Become.

1 is a diagram illustrating a configuration example of an information processing apparatus 10 according to a first embodiment. 6 is a diagram showing an example of a resource allocation table according to Embodiment 1. FIG. It is a figure which shows the example of the user information table which concerns on Embodiment 1. FIG. 6 is a flowchart illustrating an operation example of resource allocation processing according to the first embodiment. 6 is a diagram illustrating a configuration example of an information processing apparatus 100 according to Embodiment 2. FIG. FIG. 10 illustrates an example of a task management table according to the second embodiment. 12 is a flowchart illustrating an operation example of task control processing according to the second embodiment. FIG. 10 is a diagram illustrating a configuration example of an information processing apparatus 1000 according to a third embodiment. FIG. 14 illustrates an example of a task management table according to the third embodiment. 12 is a flowchart illustrating an operation example of task control processing according to the third embodiment.

Embodiment 1 FIG.
Embodiment 1 of the present invention will be described below with reference to the drawings.

  FIG. 1 is a diagram illustrating a configuration example of an information processing apparatus 10 according to the first embodiment. For example, the information processing apparatus 10 realized by a physical server or the like includes a hardware (H / W) 11 having a CPU (Central Processing Unit), a memory, a recording capacity, and the like, and a plurality of OSs (Operating Systems). A virtual software 12 for operating on the apparatus and a virtual machine 13 which is a computer having an independent OS in the physical hardware 11 divided by the virtualization technology are provided. Connect to the virtual machine 13 via the wear 11. The information processing apparatus 10 controls allocation of resources to a plurality of virtual machines 13 using the virtualization software 12. The virtualization software 12 is recorded in the recording capacity of the hardware 11 as a program, for example. Although FIG. 1 shows an example in which the virtual machine 13 configures three virtual machines as the virtual machines 13A to 13C, the number is not limited to this number.

  The virtualization software 12 includes a user management unit (user information management unit) 121, a resource control unit 122, and a resource management unit 123.

  The user management unit 121 detects whether or not the user 14 is connected to the virtual machine 13 and manages information of the connected user 14 as, for example, a user information table. The connection referred to here does not simply indicate access to the virtual machine 13, but refers to access after transmitting user information to the user management unit 121. For example, login and logout are connected and disconnected, respectively. It corresponds to. In the following description, login and logout will be described as an example, but the present invention is not limited to this.

  The resource control unit 122 controls resources by assigning resources to the plurality of virtual machines 13. Here, the resource may be at least one or a combination of two or more of the virtual CPU, virtual memory, recording capacity, and network.

  Among the resources held by the hardware 11, the resource management unit 123 relates to resources assigned to the plurality of virtual machines 13 and resources remaining in the hardware 11 after being assigned to the virtual machines 13 (surplus resources). This information (resource allocation information) is managed as a resource allocation table, for example.

  FIG. 2 is a diagram illustrating an example of a resource allocation table according to the first embodiment. The column 21 stores the name of the activated virtual machine 13. A column 22 is a CPU assigned to the virtual machine 13 and stores a value obtained by assigning a physical CPU to the virtual CPU such as the number of cores. A column 23 stores the amount of memory allocated to the virtual machine 13 and the value allocated to each virtual machine using the memory amount held by the hardware 11 as virtual memory. The column 24 stores recording areas to be assigned to the virtual machines 13, and the values assigned to the virtual machines 13 that are recording areas held by the hardware 11 are stored. A column 25 stores network bandwidths assigned to the virtual machines 13 and values assigned to the virtual machines 13 by dividing the network bandwidth held by the hardware 11.

  In the example shown in FIG. 2, the virtual machine 13A that is activated is assigned 3Core virtual CPU, 2GB virtual memory, 4TB recording capacity, and 1000Mbps network. Considering the virtual machines 13B and C in the same way, the surplus resources remaining in the hardware 11 are 2 Core for the virtual CPU, 1 GB for the virtual memory, 1 TB for the recording capacity, and 400 Mbps for the network. Therefore, for example, regarding the virtual CPU, the CPU held by the entire hardware 11 before being assigned to the virtual machine 13 is 8 Core.

  FIG. 3 is a diagram illustrating an example of a user information table according to the first embodiment. The user information table stores user information logged into the virtual machine 13 and resource information to be added.

  The column 31 stores the name of the user 14 who has logged into the virtual machine 13. The column 32 stores user information related to the user of the column 31 and includes a column 33, a column 34, and a column 35. The column 33 stores the type of the user 14 who has logged into the virtual machine 13, for example, “administrator” or “general”. The column 34 stores the priority of the user 14 who has logged into the virtual machine 13. For example, “high” may be set when the type is an administrator, and “high” and “low” may be divided according to the level of the user when the type is general. The column 35 stores the name of the logged-in virtual machine 13.

  The column 36 stores resource information that is additionally allocated to the virtual machine 13 when the user 14 logs in to the virtual machine 13, and includes a column 37, a column 38, and a column 39. A column 37 stores values of virtual CPUs to be additionally assigned to the virtual machine 13. A column 38 stores virtual memory values that are additionally allocated to the virtual machine 13. A column 39 stores a network bandwidth additionally allocated to the virtual machine 13.

  Next, a resource allocation operation will be described. FIG. 4 is a flowchart showing an operation example of the resource allocation processing according to the first embodiment.

  First, in constructing the virtual machine 13, resources are statically allocated from the hardware 11 to each virtual machine 13 (step ST401). The amount of resources allocated at this time is determined according to the application executed on the virtual machine 13 and the processing load. The allocated resources are stored in the resource allocation table shown in FIG.

  Next, the user management unit 121 detects whether or not the user 14 is connected to the virtual machine 13, that is, login or logout (step ST402). When login is detected in step ST402, the process after step ST403 is performed. On the other hand, when logout is detected, the process after step ST408 is performed.

  First, a case where login is detected in step ST402 will be described. When the login of the user 14 is detected by the user management unit 121, resource allocation is dynamically performed for the logged-in virtual machine 13. First, it is confirmed by the resource management unit 123 whether or not surplus resources of the hardware 11 exist (step ST403).

  When there is a surplus resource of the hardware 11 (step ST403-Yes), the resource control unit 122 adds a part or all of the surplus resource of the hardware 11 to the logged-in virtual machine 13 (step ST404). At this time, the amount of resources to be added is determined according to the priority of the user stored in the user information table managed by the user management unit 121. For example, when the “administrator” who is the user “a” logs in to the virtual machine A, as shown in FIG. 3, the virtual CPU is added with 2 Core, the virtual memory is 256 MB, and the network is added with 100 Mbps. By doing so, resources can be preferentially allocated to the user 14 with high priority, and processing delay due to lack of resources can be prevented. Here, the amount of additional resources is determined according to the priority of the user, but is not necessarily limited thereto, and may be determined according to the type of the logged-in virtual machine 13, for example. For example, when the priority of the virtual machine 13A is higher than that of the virtual machine 13B, the amount of additional resources is greater when the user 14 logs in to the virtual machine 13A than when the user 14 logs in to the virtual machine 13B.

  When there is no surplus resource of the hardware 11 in step ST403 (step ST403-No), the resource management unit 123 holds the surplus held by the virtual machines 13 other than the logged-in virtual machine 13 among the plurality of virtual machines 13. It is confirmed whether there is a resource (step ST405). The resource management unit 123 acquires the resource usage status of each virtual machine 13 and calculates a surplus resource by subtracting the used resource amount from the allocated resource amount stored in the resource allocation table.

  If there is a surplus resource in another virtual machine 13 (step ST405-Yes), that is, if the surplus resource calculated by the resource management unit 123 exceeds the additional resource set in the user information table, resource control Unit 122 adds a part or all of the surplus resource to the logged-in virtual machine 13 (step ST406). At this time, the surplus resource amount to be added is determined according to the priority of the user stored in the user information table held by the user management unit 121. However, the amount of allocated resources may be changed to a smaller amount than when surplus resources are allocated from the hardware 11.

  On the other hand, when there is no surplus resource in the other virtual machine 13 (step ST405 No), that is, when the surplus resource calculated by the resource management unit 123 is less than the additional resource set in the user information table, The resource control unit 122 does not add resources to the logged-in virtual machine 13 (step ST407).

  When logout is detected in step ST402, the resource management unit 123 acquires the resource usage rate of each virtual machine 13 and checks whether there is a virtual machine 13 with a high resource usage rate (step ST408). The resource usage rate can be calculated by the resource management unit 123 acquiring the resource usage status of each virtual machine 13 and calculating the used resource amount with respect to the allocated resource amount obtained from the information in the resource allocation table. .

  When there is a virtual machine 13 with a high resource usage rate (step ST408-Yes), the resource control unit 122 adds the resource added for the user 14 who logged out of the virtual machine 13 to the virtual machine 13 with a high resource usage rate. Assign (step ST409).

  On the other hand, when there is no virtual machine 13 with a high resource usage rate (step ST408-No), the resource control unit 122 allocates the resource added for the user who logged out of the virtual machine 13 to the hardware 11 (step ST410). ).

  When one of the processes of step ST404, step ST406, step ST407, step ST409, and step ST410 is performed, the resource management unit 123 updates the resource allocation table and the user information table (step ST411).

  After the processing of step ST404 is performed, the resource management unit 123 adds the information of the user 14 who has logged in to the virtual machine 13 to the user information table, the newly logged-in resource of the virtual machine 13, and the hardware 11 Update the resource allocation table for surplus resources. For example, when the resource allocation table before step ST404 is in the state shown in FIG. 2 and the user a subsequently logs in to the virtual machine 13A, the resource allocation table updated by the resource management unit 123 after step ST404 is the virtual CPU. Is 5Core, the virtual memory is 2.256 GB, and the network is 1100 Mbps. On the other hand, the surplus resources of the hardware 11 in the updated resource allocation table are 0 Core for the virtual CPU, 744 MB for the virtual memory, and 300 Mbps for the network.

  Similarly, after the process of step ST406 is performed, the resource management unit 123 adds the information of the user 14 logged in to the virtual machine 13 to the user information table, the resource of the logged-in virtual machine 13, and the logged-in virtual The resource allocation table is updated for resources of other virtual machines 13 that have provided surplus resources to the machine 13.

  After the process of step ST407 is performed, the information of the user 14 who has logged into the virtual machine 13 is added to the user information table. Information about the resource allocation table is not updated.

  After the process of step ST409 is performed, the information of the user 14 logged out from the virtual machine 13 is deleted from the user information table, and the resource of the logged out virtual machine 13 and the resource added for the logged out user are newly added. The resource allocation table is updated for the resources of the virtual machine 13 allocated to.

  After the processing of step ST410 is performed, the information of the user 14 logged out from the virtual machine 13 is deleted from the user information table, and the resource information of the logged out virtual machine 13 and the resource added for the logged out user are added. The resource allocation table is updated for the surplus resources of the newly allocated hardware 11.

As described above, according to the first embodiment of the present invention, the login of the user 14 to the virtual machine 13 is detected, and resources are added to the logged-in virtual machine 13 based on the information of the logged-in user 14. Thus, it is possible to allocate sufficient resources before performing task execution processing, and it is possible to prevent processing delay due to resource shortage.

  In addition, since the amount of resources to be added is determined according to the priority of the logged-in user 14, resources can be allocated with priority to the user 14 with a higher priority, and resource shortage can be prevented more reliably.

  Further, the resources to be additionally allocated to the logged-in virtual machine 13 are allocated from the hardware 11 when there is a surplus resource of the hardware 11, and are allocated from the surplus resources of another virtual machine 13 when there is no surplus resource of the hardware 11. Since it did in this way, it is not necessary to use the surplus resource of the other virtual machine 13 as much as possible, and even if a sudden task occurs in the other virtual machine 13, a resource shortage can be prevented.

  In addition, after the user 14 logs out, the resources added for the logged-in user 14 are allocated to the other virtual machines 13, so that the resources of the entire physical server can be efficiently distributed to the plurality of virtual machines 13. Is possible.

  As described above, when there is no surplus resource of the hardware 11, it is preferable to allocate an additional resource from another virtual machine 13, but this is not always necessary. For example, it may be configured to first check whether there is a surplus resource of another virtual machine 13 and to allocate from the surplus resource of the hardware 11 if it does not exist. Further, it is not always necessary to allocate resources from only one of the hardware 11 and the other virtual machine 13, and resources may be allocated from both of them. By doing this, for example, when a user 14 with a high priority logs in, only the surplus resource of the hardware 11 is not sufficient for the additional resource, and the surplus resource of the other virtual machine 13 is only the additional resource. Even if it is not enough, if the additional resources are satisfied by summing up these surplus resource amounts, additional resources can be allocated to the virtual machine 13 to which the user 14 has logged in, resulting in insufficient resources. Therefore, it is possible to prevent the processing delay due to.

  In addition, in step ST405 of FIG. 4, when the other virtual machine 13 has the surplus resource, the configuration for adding the surplus resource of the other virtual machine 13 to the logged-in virtual machine 13 has been described, but the present invention is not limited thereto. For example, when there is a virtual machine 13 with a low resource usage rate, the resource of the virtual machine 13 may be added to the logged-in virtual machine 13. In addition, when there are a plurality of other virtual machines 13 having surplus resources, the resources of the virtual machine 13 having surplus resources and low resource usage rate are added to the logged-in virtual machine 13. Also good.

Embodiment 2. FIG.
The second embodiment of the present invention will be described below with reference to the drawings. In the second embodiment, when the resources of tasks executed by the virtual machine 13 are insufficient, necessary resources are secured by stopping some of the plurality of tasks executed by the virtual machine 14. To prevent a shortage of resources for the task to be executed.

  FIG. 5 is a diagram illustrating a configuration example of the information processing apparatus 100 according to the second embodiment. In the information processing apparatus 100 according to the second embodiment, the virtualization software 12 newly includes a task execution control unit 124 and a task management unit 125 in addition to the configuration of the information processing apparatus 10 according to the first embodiment. However, this is different from the first embodiment. In the example of FIG. 5, the virtual machine 13A executes the plurality of tasks 131A to 131C among the plurality of virtual machines 13A to 13C constructed by the information processing apparatus 100, but is not limited thereto. The tasks are executed in the virtual machines 13B and 13C as well as the virtual machine 13A, but the description is omitted for simplicity. The same applies to the following description. Note that the tasks 131A to 131C may be the same application or different applications. Since the other configuration of the information processing apparatus 100 according to the second embodiment is the same as that of the first embodiment, the same reference numerals as those in FIG.

  The task execution control unit 124 has the authority to execute the task 131 that operates on the virtual machine 13, receives an execution instruction for the task 131 from the virtual machine 13, and instructs the virtual machine 13 to execute or stop the task 131. To do.

  The task management unit 125 manages information (task management information) such as the type, priority, and used resource amount of the task 131 operating on the virtual machine 13 as a task management table, for example.

  FIG. 6 is a diagram showing an example of a task management table according to the second embodiment. The column 61 stores the name of the target virtual machine 13. In the example of FIG. 6, information about tasks that operate on the virtual machine 13A is stored. The column 62 stores the name of the task 131 operating on the virtual machine 13A. The column 63 stores the type of the task 131 operating on the virtual machine 13A. The types include, for example, “user” in which an operation command is issued by the user 14, “scheduler” in which an operation command is performed by scheduling such as maintenance, and “normal operation” that needs to operate constantly on the system.

  The column 64 stores the priority of the task 131 operating on the virtual machine 13A. The priority is determined mainly by the type stored in the column 63. The column 65 stores the execution state of the task 131 operating on the virtual machine 13A. For example, the execution state is “execution” as the current execution state, “stop” when the execution cannot be executed for the reason of the execution being attempted, and “start” as the start state. Here, a task 131 whose execution state is “execution” is defined as an execution task, a task 131 whose execution state is “stop” is defined as a stop task, and a task 131 whose execution state is “start” is defined as a start task.

  A column 66 stores a measure of the CPU usage of the task 131 operating on the virtual machine 13A. This value may be set automatically by the system once the task 131 is operated, or may be set manually by the administrator. A column 67 stores a measure of the memory usage of the task 131 operating on the virtual machine 13A. This value may be set automatically by the system once the task 131 is operated, or may be set manually by the administrator. A column 68 stores an indication of the network usage of the task 131 operating on the virtual machine 13A. This value may be set automatically by the system once the task 131 is operated, or may be set manually by the administrator.

  Next, the task control operation will be described. FIG. 7 is a flowchart showing an operation example of the task control process according to the second embodiment.

  First, the task management unit 125 acquires task management information for the task 131 operating in the virtual machine 13, and stores the information in the task management table (step ST701).

  Next, the task execution control unit 124 checks whether or not an execution instruction for the start task 131 has occurred in the virtual machine 13 (step ST702). Hereinafter, as an example, the task 131a operating on the virtual machine 13A will be described as a start task.

  If an execution instruction for the start task 131A is generated in step ST702, the process after step ST703 is performed, and if an execution instruction for the start task 131A is not generated, the process after step ST710 is performed.

  First, the process after step ST703 is demonstrated. When an execution instruction for the start task 131A is generated (step ST702-Yes), the task management unit 125 stores information on the start task 131A (start task a) in the task management table. Further, the resource management unit 123 notifies the task management unit 125 of information about the surplus resources of the virtual machine 13A, and the task management unit 125 lacks resources necessary for executing the start task 131A in the virtual machine 13A. Is determined (step ST703). The task management unit 125 compares the surplus resource amount of the virtual machine 13A notified from the resource management unit 123 with the resource amount necessary to execute the start task 131A stored in the task management table, thereby Determine if there is a shortage. In the example of FIG. 6, the amount of resources such as CPU and memory required for executing the task 131 is “medium” and “high”, but more specifically, information on the necessary amount of resources is quantified. Stored.

  If the task management unit 125 determines that the resource for executing the start task 131A is insufficient (step ST703-Yes), the resource management unit 123 determines whether there is a surplus resource in the hardware 11 based on the information in the resource management table. It is confirmed whether or not (step ST704).

  When the resource management unit 123 determines that there is no surplus resource in the hardware 11 (step ST704-No), the task management unit 125 determines the priority of the start task 131A and other executions executed in the virtual machine 13A. The priority of the task 131 is acquired from the task management table, and it is confirmed whether or not there is an execution task 131 having a lower priority than the start task 131A (step ST705).

  When the task management unit 125 determines that there is an execution task 131 having a lower priority than the start task 131A (step ST705-Yes), the task execution control unit 124 stops the execution task 131 having a lower priority and performs the task The management unit 125 moves the execution task 131 to the stop task 131 (step ST706).

  On the other hand, when the task management unit 125 determines that there is no execution task 131 having a lower priority than the start task 131A (No in step ST705), resources necessary for executing the start task 131A are assigned to the virtual machine 13A. Since it cannot be secured, the task management unit 125 shifts the start task 131A to a stop task (step ST707). Taking the case shown in FIG. 6 as an example, there is no execution task 131 having a lower priority than the start task 131A in the virtual machine 13A.

  In step ST704, when there is a surplus resource in the hardware 11 (step ST704-Yes), the resource control unit 122 allocates a part or all of the surplus resource of the hardware 11 to the virtual machine 13A that executes the start task 131A. It is allocated as an additional resource (step ST708). The resource amount to be added at this time may be a resource amount that can at least execute the start task 131A.

  When the execution task 131 having a low priority is transferred to the stop task 131 in step ST706, a resource is added to the virtual machine 13A that executes the start task 131A in step ST708, or a resource for executing the start task 131A is found. If there is no shortage in the virtual machine 13A (step ST703-No), the task management unit 125 moves the start task 131A to the execution task 131, and the task execution control unit 124 allows the virtual machine 13A to execute the task 131. Is issued, that is, the task 131 is instructed to be executed (step ST709).

  Next, the processing after step ST710 when the execution instruction of the start task 131A is not generated (step ST702-No) will be described.

  The task management unit 125 checks whether there is a stopped task 131 based on the information stored in the task management table. If the stopped task 131 exists, the task management unit 125 notifies the hardware 11 notified from the resource management unit 123. It is confirmed whether the stop task 131 can be executed using the surplus resources (step ST710).

  When the task management unit 125 determines that the stop task 131 can be executed by using the surplus resource of the hardware 11 (step ST710-Yes), the task management unit 125 moves the stop task 131 to the execution task 131, and the task The execution control unit 124 instructs the virtual machine 13A to execute the task 131 (step ST711).

  As described above, according to the second embodiment of the present invention, when the start task 131A is executed in the virtual machine 13A, the other execution tasks 131 are controlled to stop according to the priority of the task 131. It is possible to execute a task 131 with a high priority after securing sufficient resources, and to prevent a processing delay due to lack of resources.

  Further, when there are surplus resources in the hardware 11, the surplus resources of the hardware 11 are additionally allocated to the virtual machine 13A that executes the start task 131A. Therefore, the start task 131A is secured after securing sufficient resources. Can be executed.

  In addition, the task 131 once stopped due to a resource shortage in the virtual machine 13A can be re-executed after securing sufficient resources by additionally allocating surplus resources from the hardware 11.

  Heretofore, for example, in steps ST704 and ST708 of FIG. 7, the process of allocating an additional resource to the virtual machine 13A that executes the start task 131A when there is a surplus resource in the hardware 11 has been described. However, as in the first embodiment, the surplus resources of other virtual machines 13 other than the virtual machine 13A may be added, or the surplus resources of the other virtual machines 13 and the surplus resources of the hardware 11 are added. You may make it do.

Embodiment 3 FIG.
Embodiment 3 of the present invention will be described below with reference to the drawings. In the third embodiment, when the resources of tasks executed by the virtual machine 13 are insufficient, the resources for some of the plurality of tasks executed by the virtual machine 14 are limited, so that it is necessary. Reserve resources to prevent resource shortages for tasks to be executed.

  FIG. 8 is a diagram illustrating a configuration example of the information processing apparatus 1000 according to the third embodiment. The information processing apparatus 1000 according to the third embodiment is different from the information processing apparatus according to the first and second embodiments in that the virtualization software 12 newly includes a task resource control unit 126. Different from 1 and 2. Since other configurations are the same as those of the first and second embodiments, the same reference numerals as those in FIGS.

  The task resource control unit 126 limits the resource amount of a task that operates on the virtual machine 13 based on information on resource limitation stored in the task management table managed by the task management unit 125.

  FIG. 9 is a diagram illustrating an example of a task management table according to the third embodiment. The task management table of FIG. 9 differs from the task management table described with reference to FIG. 6 in that information regarding resource restrictions is newly added and stored. Since other configurations are the same, the same reference numerals as those in FIG.

  The column 9 is composed of a column 91 and a column 92 and stores the resource limit amount of the task 131. A column 91 stores a value about a limit width of an I / O (Input / Output) resource among resources used in the task 131 operating on the virtual machine 13. The limit width of the I / O resource can be set by the administrator. The column 92 stores a value for the limit width of the network resource among the resources used in the task 131 operating on the virtual machine 13. The limit width of the network resource can also be set by the administrator.

  Next, the task control operation will be described. FIG. 10 is a flowchart showing an operation example of the task control process according to the third embodiment. Note that the processing in steps ST701 to ST711 in FIG. 10 is the same as the processing in steps ST701 to ST711 described with reference to FIG.

  In step ST705, when there is an execution task 131 having a lower priority than the start task 131A in the virtual machine 13A (step ST705-Yes), the resource limit calculated by the task management unit 125 based on the information stored in the task management table Whether or not the start task 131A can be executed based on the predicted value of the resource of the subsequent execution task 131 and the surplus resource amount of the virtual machine 13A notified from the resource management unit 123, that is, the start task 131A is executed. It is confirmed whether or not a resource for doing so is secured (step ST1001).

  When the start task 131A can be executed by limiting the resources of the execution task 131 having a low priority (step ST1001-Yes), the task resource control unit 126 stores it in the task management table managed by the task management unit 125. Based on the information regarding the resource limitation, the resource of the execution task 131 having a low priority is limited (step ST1002). After the process of step ST1002, the process of step ST709 is performed next.

  On the other hand, if the start task 131A cannot be executed even if the resource of the execution task 131 with a low priority is limited in step ST1001, the task resource control unit 126 performs the resource limitation process for the execution task 131 with a low priority. Is not performed, and the process of step ST706 is performed.

  In step ST705, when there is no execution task 131 having a lower priority than the start task 131A in the virtual machine 13A (step ST705-No), the resource limit calculated by the task management unit 125 based on the information stored in the task management table Whether the start task 131A can be executed based on the predicted value of the resource of the subsequent start task 131A and the surplus resource amount of the virtual machine 13A notified from the resource management unit 123, that is, the start task 131A is executed. It is confirmed whether or not a resource for doing so is secured (step ST1001). That is, unlike step ST1001, the task management unit 125 checks whether or not resources for executing the start task 131A can be secured in the virtual machine 13A by limiting the resources of the start task 131A itself.

  When the start task 131A can be executed by limiting the resources of the start task 131A (step ST1003-Yes), the task resource control unit 126 relates to the resource limit stored in the task management table managed by the task management unit 125. Based on the information, the resource of the start task 131A is limited (step ST1004). After the process of step ST1004 is performed, the process of step ST709 is performed next.

  On the other hand, if the start task 131A cannot be executed even if the resource of the start task 131A is limited in step ST1003 (step ST1003-No), the task resource control unit 126 does not limit the resource of the start task 131A, Processing is done.

  As described above, according to the third embodiment of the present invention, when the start task 131A is executed in the virtual machine 13A, the resources of other execution tasks 131 having lower priority than the start task 131A are limited. Therefore, it is possible to secure sufficient resources to execute the start task 131A without stopping other execution tasks 131, and it is possible to prevent a processing delay due to resource shortage.

  In addition, when there is no execution task 131 having a lower priority than the start task 131A, the start task 131A is executed by limiting the resources of the start task 131A itself, so that the other execution tasks 131 are stopped. In addition, since the start task 131A can be executed without being disabled, the resources allocated to the virtual machine 13 can be used efficiently.

  In the present invention, within the scope of the invention, any combination of the embodiments, or any modification of any component in each embodiment, or omission of any component in each embodiment is possible. .

  10, 100, 1000 Information processing apparatus, 11 hardware, 12 virtualization software, 13A to C virtual machines, 14 users, 121 user management unit, 122 resource control unit, 123 resource management unit, 124 task execution control unit, 125 Task management unit, 126 Task resource control unit

Claims (10)

An information processing apparatus that controls resource allocation to a plurality of virtual machines,
A user information management unit that detects whether a user is connected to the virtual machine and manages information of the connected user;
Among the hardware resources, a resource management unit that manages the resources allocated to the plurality of virtual machines and the hardware surplus resources as resource allocation information;
A resource control unit that allocates resources to the plurality of virtual machines,
When the user information management unit detects a connection to the first virtual machine by the user, the resource control unit allocates resources to the first virtual machine based on information of the connected user. Information processing apparatus.   The resource control unit includes a surplus resource held by any other virtual machine other than the first virtual machine among the plurality of virtual machines, a surplus resource of the hardware, or a surplus held by the other virtual machine. The information processing apparatus according to claim 1, wherein resources of both a resource and a surplus resource of the hardware are allocated to the first virtual machine.   The information processing apparatus according to claim 2, wherein the resource control unit allocates a part or all of the hardware surplus resources to the first virtual machine if surplus resources exist in the hardware. .   The resource control unit, when there is no surplus resource in the hardware and when there is a surplus resource in the other virtual machine, assigns the surplus resource of the other virtual machine to the first virtual machine. The information processing apparatus according to claim 2, wherein the information processing apparatus is assigned to a machine.   When the user control unit detects that the user has lost connection to the first virtual machine, the resource control unit allocates the resource allocated to the first virtual machine to the other virtual machine or the 5. The information processing apparatus according to claim 2, wherein the information processing apparatus is assigned to hardware. The resource management unit selects a virtual machine with a high resource usage rate from the plurality of virtual machines based on the resource management information,
The information processing apparatus according to claim 5, wherein the resource control unit allocates the resource allocated to the first virtual machine to the virtual machine selected by the resource management unit.   The information according to any one of claims 1 to 6, wherein the user information management unit detects whether or not there is a connection to the virtual machine based on a login or logout to the virtual machine by the user. Processing equipment. A task management unit for managing task management information including at least information on priority of a task executed in the virtual machine;
A task execution control unit that instructs the virtual machine to execute or stop the task,
The task execution control unit, based on the task management information, when the second virtual machine among the plurality of virtual machines does not have a resource necessary for executing the first task, The information processing apparatus according to claim 1, wherein an instruction is given to stop a second task having a lower priority than the task.   The resource control unit, when the second virtual machine does not have a resource necessary for executing the first task, a surplus resource held by the other virtual machine or a surplus of hardware 9. The information processing apparatus according to claim 8, wherein a resource, or both of a surplus resource held by the other virtual machine and a surplus resource of the hardware are allocated to the second virtual machine. A task resource control unit that limits the resources used to execute a task,
The task execution control unit allows the second virtual machine to execute the first task even if the resource used by the task resource control unit to execute the second task is limited. 10. The information processing apparatus according to claim 8, wherein the second task is instructed to stop when the necessary resource is not included.

Images