JP2013117806A - Cpu allocation time management device and cpu allocation time management program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device for executing an appropriate control according to a function request when there is CPU free time, in a device for executing a CPU allocation time management system for controlling the upper limit of the CPU allocation time for each task group.SOLUTION: A group 1 comprises tasks 1a, 1b, 1c, 1d. A group 2 comprises tasks 2a, 2b, 2c, 2d. A CPU allocation time management device 1000 ends execution of only the tasks 1a, 1b at the time (a) after the lapse of the CPU allocation time set for the group 1 in a CPU time management cycle T3. Further, the CPU allocation time management device 1000 ends execution of the tasks 2a-2d at the time (b) leaving the residual time before the lapse of the CPU allocation time set for the group 2 in the CPU time management cycle T3. The CPU allocation time management device 1000 executes the tasks 1a, 1b not executed yet in the residual time for the group 2 in the CPU time management cycle T3.

Description

  The present invention relates to a CPU allocated time management method for setting an upper limit of CPU time that can be used by a task group composed of a plurality of tasks in an OS executing a plurality of tasks.

  In the case of an information processing apparatus that executes a plurality of different functions simultaneously (multitask), it is necessary to continue the execution of each function with predictable performance. In order to realize such performance predictability, a task group is composed of task groups that constitute functions, and the upper limit of the CPU time that can be used by the task group is limited. There is a CPU allocation time management method that limits use of time. For example, in one conventional CPU allocation time management method, a virtual machine is configured for each function, a service rate indicating the upper limit of the CPU allocation time is set for each virtual computer, and the CPU usage time of the virtual computer is monitored to By making the ratio below, one function does not occupy all CPU time.

  Furthermore, in order to prevent the CPU time from being allocated for a long time due to reaching the upper limit of the CPU allocation time, and to prevent the response of the function from being greatly delayed, the increase rate of the CPU usage time of the virtual machine is measured, Control was performed to distribute CPU idle time within the allocated time (for example, Patent Document 1). In addition, there is a problem that if the upper limit of the CPU allocation time in a short period is limited for functions (task groups, virtual machines) that are temporarily in a high load state, the CPU usage time may be exceeded. . In response to this problem, another method enables selection of CPU allocation time management in a short period and CPU allocation time management in a long period of time, and processing that is temporarily high in load is managed in a long period of time. Has been used to allow a temporary CPU allocation time to be exceeded (for example, Patent Document 2).

JP 2008-186210 A Japanese Patent Laid-Open No. 9-81401

  However, although the conventional technology can control the upper limit of the CPU time used by each function, when each function does not use up the CPU allocated time and the CPU has a free time, the CPU free time is used upon request. There was a problem that it could not be controlled as possible. If it is set not to perform CPU allocation time management for a specific task group, the task group can freely use the CPU idle time. However, in this case, even when the CPU load is high, there is a problem that the upper limit of the CPU time of the task group cannot be limited, which affects the CPU time of other task groups. Therefore, a system for controlling the allocation of the CPU free time after performing the CPU allocation time management is required.

  The present invention is a device that executes a CPU allocation time management system that performs upper limit control of CPU allocation time for each task group (for each processing unit in a cycle), and is suitable for function requests when there is CPU idle time. It is an object of the present invention to provide an apparatus that executes a CPU allocated time management system that performs various controls.

The PU allocation time management device of the present invention is:
CPU allocated time indicating the upper limit value of the continuous use time of the CPU, which is a processing unit of the CPU within the execution cycle scheduled for each execution cycle set for the CPU (Central Processing Unit). A first execution unit that causes a CPU to execute an intra-period processing unit that is a processing unit for which the CPU is set within a CPU allocation time;
When a plurality of in-cycle processing units are executed by the CPU in any execution cycle, an in-cycle unfinished unit that is an in-cycle processing unit in which execution by the CPU does not end occurs during the CPU allocation time, An in-cycle processing unit that is executed after an in-cycle unfinished unit during the execution cycle in which the in-cycle unfinished unit is generated, and that is executed within a period shorter than the CPU allocation time. And a second execution unit that causes the CPU to execute the unfinished unit within the cycle following the end of the execution of the early termination unit within the cycle when the early termination unit within the cycle occurs.

  According to the present invention, an apparatus that executes a CPU allocation time management system that performs upper limit control of CPU allocation time for each task group, and that performs appropriate control in response to a function request when there is CPU idle time An apparatus for executing a management method can be provided.

1 is a block diagram of a CPU allocation time management apparatus 1000 according to Embodiment 1. FIG. 5 is a flowchart of initial setting performed by the OS 101 according to the first embodiment. FIG. 6 illustrates an operation in a hardware restriction mode according to the first embodiment. FIG. 6 is a diagram for explaining an operation in a soft restriction mode in the first embodiment. 4 is an operation flow of the CPU scheduling function unit 111 according to the first embodiment. The operation | movement flow of the free resource management function part 121 in Embodiment 1. FIG. FIG. 10 is a block diagram of a CPU allocation time management apparatus 1000 according to the second embodiment. FIG. 10 is a block diagram of a CPU allocation time management apparatus 1000 according to the third embodiment. FIG. 14 is an external view of a CPU allocation time management apparatus 1000 according to the fourth embodiment. FIG. 10 is a hardware configuration diagram of a CPU allocation time management apparatus 1000 according to the fourth embodiment.

Embodiment 1 FIG.
FIG. 1 is a configuration diagram showing a CPU allocation time management apparatus 1000 that executes the CPU allocation time management method. FIG. 1 is as follows.
(1) 101 is an OS (execution cycle setting unit).
(2) The task groups 201 and 202 are task groups (also simply referred to as groups) that are units for collecting the tasks operating on the OS 101 for each function and managing the CPU allocation time. Each group (an example of an intra-cycle processing unit) includes one or a plurality of tasks. There may be two or more groups and a plurality of groups.
(3) The CPU scheduling function unit 111 (first execution unit) performs CPU scheduling in the OS 101.
(4) The CPU allocation time table 112 is a table for the CPU scheduling function unit 111 to record CPU allocation time information for each group. The CPU allocation time table 112 records the CPU allocation time which is the CPU time limit for each group and the CPU usage time using the CPU within the management cycle time (within the CPU time management cycle). The CPU allocation time indicates the upper limit value of the continuous use time of the CPU that is permitted within the management cycle time for the task group. Although FIG. 1 shows an example in which the CPU time is shown as a percentage in the management cycle time, it may be recorded in actual time units.
(5) The scheduling queue 113 is a scheduling queue for the CPU scheduling function unit 111 to select an executable task. What has been described so far is a component also in the conventional system.

In the first embodiment, in addition to the above configuration, the following components are newly added.
(1) The free resource management function unit 121 (second execution unit) performs allocation control of CPU free time.
(2) The free resource pool 122 is a resource pool in which the free resource management function unit 121 reserves CPU time. The free resource pool 122 includes idle tasks as tasks that consume CPU time in an infinite loop.
(3) The resource limit mode 123 is a mode for setting a free CPU time allocation policy. In the resource limit mode 123, it is possible to set a “hard limit mode” in which no allocation is made to a group even if there is a free CPU time, and a “soft limit mode” in which the allocation is performed when there is a free CPU time.
(4) The free resource available group setting 124 registers a group (unfinished execution target) that can use the free CPU time when the “soft restriction mode” is set. A plurality of groups may be registered in the free resource available group setting 124.
(5) The CPU time management cycle setting 125 registers a CPU time management cycle (execution cycle) that is a unit cycle for managing the CPU allocation time. The CPU scheduling function unit 111 and the free resource management function unit 121 indicate that the CPU allocation time, the CPU usage time, and the CPU free time become as allocated in the time (cycle) registered in the CPU time management cycle setting 125. to manage.
(6) The allocation overtask queue 126 is a queue for registering a task that requests the CPU beyond the CPU allocation time.

Next, the operation will be described.
FIG. 2 shows the initial setting operation of the CPU allocation time management method performed by the OS 101. The subject of the operation in FIG.
(1) First, in S101, group groups 201, 202,... Serving as a unit of CPU allocation time management are created in the same manner as in the conventional CPU allocation time management method. The CPU allocation time for the group created here is registered in the corresponding entry of the CPU allocation time table 112.
(2) Next, in S102, an idle task is created as a task for free CPU time reservation, and is registered in the free resource pool 122. Further, in order to manage the CPU allocation time for the free resource pool 122, the free resource pool 122 is registered in the CPU allocation time table 112.
(3) Next, in S103, the initial value of the resource restriction mode 123 is set. If CPU scheduling is not performed for a group that exceeds the CPU allocation time even when there is an empty CPU time, set to “hard limit mode”. When CPU scheduling is performed, the “soft limit mode” is set.
(4) Next, in S104, a group group that permits the use of the free CPU time is set in the free resource available group setting 124. A plurality of groups permitting the use of the free CPU time may be registered. This setting has no effect in the “hard limit mode”.
(5) Next, in S105, a unit cycle time (execution cycle) for managing the CPU allocation time is set in the CPU time management cycle setting 125. This time is a time period for checking the upper limit of the CPU allocation time, accumulating the CPU usage time, and checking whether there is a CPU idle time. By reducing this time, the CPU allocation time control is performed in a short cycle. With these initial setting operations, the CPU allocation time management parameters of the free resource management function unit 121 are initialized, and the CPU scheduling function unit 111 and the free resource management function unit 121 operate based on the settings. .

  Next, the “hard limit mode” and the “soft limit mode” of the resource limit mode 123 will be described with reference to FIGS. 3 and 4 are diagrams showing the execution status of the group, and the horizontal axis represents the CPU time. This example shows a case where two task groups, group 1 and group 2, are created.

(Hard limit mode)
FIG. 3 shows a case where the resource limit mode is set to “hard limit mode”. The CPU allocation time is managed based on the CPU time management period set in the CPU time management period setting 125. In the first period T1, both the group 1 and group 2 CPU allocation times are used up, but in the second period T2, group 1 does not use up the CPU allocation time. However, since the resource limit mode is set to the “hard limit mode”, the CPU free time is reserved by the idle task of the free resource pool without being assigned to another group. Further, in the third period T3, the group 2 does not use up the CPU allocation time, but similarly, the CPU idle time is reserved by the idle task without being allocated to other groups. As a result, both the group 1 and the group 2 can realize the uniformity of the execution performance by the upper limit of the CPU allocation time.

(Soft restriction mode)
FIG. 4 shows a case where the resource limit mode is set to “soft limit mode”. Looking at the effect in the same situation as in FIG. 3, in the second period T2, group 1 does not use up the CPU allocation time, but group 2 is not set in “free resource available group setting”. The CPU idle time cannot be used and is reserved by the idle task. However, if there is a free time in the CPU allocation time of group 2 (an example of early termination unit in the cycle) in the third cycle T3, “resource limit mode” is set to “soft limit mode” and “ “Available resource available group setting” is set to “Group 1”. Therefore, the group 1 (an example of the unfinished unit in the cycle) can use the CPU idle time, and the group 1 is executed exceeding the CPU allocation time. As a result, group 2 can achieve constant execution performance, while group 1 can perform high-speed execution using CPU time equal to or greater than the CPU allocation time.

  In FIGS. 3 and 4, the group 1 and the group 2 are schematically shown so as to be scheduled sequentially and continuously. However, this is only an example, and in reality, tasks in the group 1 and group 2 can be finely scheduled with respect to task execution order and task switch granularity by the task scheduling of the OS 101.

Next, the operation of the CPU scheduling function unit 111 will be described with reference to FIG.
FIG. 5 shows the operation of the CPU scheduling function unit 111 in the CPU allocation time management method. That is, the subject of the operation in FIG. 5 is the CPU scheduling function unit 111. In the following description of FIG. 5 and FIG. 6, it is assumed that group 1 and group 2 are assumed as shown in FIG. 4, and group 1 has tasks 1a to 1d, and group 2 has tasks 2a to 2d.
(1) First, in S200, the CPU scheduling function unit 111 selects and executes a task to be executed according to the task scheduling of the OS 101. The OS has various task scheduling methods. In the CPU allocation time management method executed by the CPU allocation time management apparatus 1000, the task scheduling method is used as it is. As long as the CPU usage time of the group does not reach the CPU allocation time, the CPU allocation time management method of the CPU allocation time management apparatus 1000 does not affect the task scheduling of the OS.
(2) Next, in S201, the CPU time used by the tasks in each group is integrated and recorded in the CPU usage time of the corresponding group in the CPU allocation time table 112. As shown in FIG. 4, when the program is executed from group 1, the CPU usage time of group 1 is recorded.
(3) Next, in S202, it is confirmed whether the CPU usage time of the group reaches the CPU allocation time and there is an executable task in the group. If there is an executable task, the task is removed from the scheduling queue 113 and registered in the allocation overtask queue 126 in S203. For example, in the third period T3 in FIG. 4, the CPU allocation time for group 1 is reached, and tasks 1c and 1d are not yet executed and exist in the scheduling queue 113. In this case, the CPU scheduling function unit 111 removes the tasks 1c and 1d from the scheduling queue 113 and registers them in the allocation overtask queue 126.
(4) In S204, it is confirmed whether or not the management cycle time registered in the CPU time management cycle setting 125 has elapsed in the execution of the processing from S200 to S203 so far. Currently, the time (a) of the third cycle T3 in FIG. 4 is assumed.
(5) If the CPU time management period T has elapsed, the CPU usage time of all groups in the CPU allocation time management table 112 is reset to zero in S205, and the tasks registered in the allocation overtask queue 126 are transferred to the scheduling queue. Return to 113.
Thereafter, the process returns to S200.
(6) If the CPU time management period T has not elapsed in S204, it is confirmed in S206 whether the tasks registered in the scheduling queue 113 are only idle tasks in the free resource pool 122. If a task other than the idle task is registered in the scheduling queue 113, the process returns to S200. The process returns from S206 to S200 when the tasks 2a to 2d are registered in the scheduling queue 113 after the time (a) in FIG.
(7) If the task registered in the scheduling queue in S206 is only an idle task, the free resource management function unit 121 is called and executed in S300. Details of the operation of S300 will be described with reference to FIG. For example, this is a case where the tasks 2a to 2d have been executed as in the state of time (b) in FIG.

FIG. 6 shows details of S300 in FIG. 5 which is an operation of the free resource management function unit 121 of the CPU allocation time management apparatus 1000. 6 is the free resource management function unit 121.
(1) First, in S301, the free resource management function unit 121 checks whether the resource limit mode 123 is set to “hard limit mode” or “soft limit mode”. When the “hard limit mode” is set, the idle task of the free resource pool 122 is executed in S304 without assigning the free CPU time to another group. If the resource restriction mode is set to “soft restriction mode” in S301, the process proceeds to S302.
(2) In S302, it is confirmed whether the allocation overtask queue 126 is empty. If the allocation overtask queue 126 is empty, there is no task in an executable state, and an idle task is executed in S304.
(3) If a task is registered in the allocation overtask queue 126 in S302, among the tasks registered in the allocation overtask queue 126 in S303, the group registered in the “free resource available group setting 124” Perform the task. The method for selecting a task to be executed at this time follows the task scheduling method of the OS. In this example, it is assumed that the tasks 1c and 1d are registered in the allocation overtask queue 126 at the time (b) in FIG. 4, and the group 1 is registered in the “free resource available group setting 124”. .
(4) In S305, it is confirmed whether or not the management cycle time registered in the CPU time management cycle setting 125 has passed through the execution of the processing from S301 to S304. If the CPU management cycle T has not elapsed, the process returns to S301.
(5) If the CPU time management period T has elapsed in S305, the CPU usage time of all groups in the CPU allocation time management table 112 is reset to zero in S306, and tasks registered in the allocation overtask queue 126 are scheduled. Return to the queue 113. Thus, the operation of the free resource management function unit 121 is completed, and the process returns to S200 of FIG. For example, when the task 1d remains in the allocation overtask queue 126 in S306, the free resource management function unit 121 returns the task 1d to the scheduling queue 113.

  As described above, in the CPU allocation time management method of the CPU allocation time management apparatus 1000 according to the first embodiment, when there is CPU free time, the OS calls the free resource management function unit 121 and follows the parameters set there. The task group controls whether or not the CPU can be used beyond its CPU allocation time. As a result, the CPU free time allocation method can be changed depending on whether the function executed by the task group requires high speed or constant performance.

Embodiment 2. FIG.
A CPU allocation time management apparatus 1000 according to the second embodiment will be described with reference to FIG. In the second embodiment, the CPU allocation time management apparatus 1000 according to the first embodiment further includes a free resource management parameter setting function unit 131 (the function will be described later). According to the second embodiment, by providing the free resource management parameter setting function unit 131 (period change unit, mode switching unit, unfinished execution target change unit), the free resource management function unit 121 is used from the upper OS level. By enabling the parameter to be changed, the CPU allocation time management operation can be changed when the CPU allocation time management apparatus 1000 (information processing apparatus) is executed.

  FIG. 7 is a configuration diagram in such a case. The free resource management parameter setting function unit 131 has a function of changing a parameter used by the free resource management function unit 121 when the CPU allocation time management method is executed from the upper OS. In FIG. 7, the components having the same numbers as those in FIG. 1 are the same as those in FIG.

The free resource management parameter setting function unit 131 can change the following setting values (parameters) (a) to (c) of the free resource management function unit 121 initially set by the OS 101 by the operation of FIG. 2 in the first embodiment. It is.
That is,
(A) Resource restriction mode 123,
(B) Free resource available group setting 124,
(C) CPU time management cycle setting 125
Is changed after the initial setting in FIG. 2 by “user instruction or inter-task call”.

  In the operations of FIGS. 5 and 6 described in the first embodiment, the respective setting values (parameters (a) to (c)) are read during the operation to make a determination. By changing, the operation of the free resource management function unit 121 can be changed.

  As described above, in the CPU allocation time management method of the CPU allocation time management apparatus 1000 according to the second embodiment, by adding the free resource management parameter setting function unit 131, the parameters used by the free resource management function unit 121 from the upper level of the OS. Can be changed not only when the CPU allocation time management apparatus 1000 (information processing apparatus) is started but also when it is executed. As a result, the CPU free time allocation control of the free resource management function unit 121 can be changed according to a change in a request for a task function or a change in state.

  For example, although it is normally operated in the “hard limit mode”, the free resource management parameter setting function unit 131 changes the setting to the “soft limit mode” only when “a process different from normal operation” occurs. Thus, it is possible to perform control such that free CPU time is allocated to a task group that executes a function related to “a process different from normal processing at the time of activation”.

Embodiment 3 FIG.
A CPU allocation time management apparatus 1000 according to the third embodiment will be described with reference to FIG. In the CPU allocation time management apparatus 1000 according to the first embodiment, the CPU allocation time management method is applied to the OS, and CPU allocation time management between tasks on the OS is realized. In this third embodiment, the CPU allocation time management method of the first embodiment is applied to a hypervisor that executes a plurality of OSs (or a plurality of virtual machines), and between OSs (or between virtual machines) on a hypervisor. Realizes CPU allocation time management. The third embodiment can be combined with the second embodiment in the same manner as the first embodiment.

  FIG. 8 is a configuration diagram of the CPU allocation time management apparatus 1000 according to the third embodiment. In the CPU allocation time management apparatus 1000 according to the third embodiment, the free resource management function unit 121 of the CPU allocation time management method is incorporated in the hypervisor 501 instead of the OS 101 shown in the first embodiment.

  The hypervisor 501 performs hardware management and virtualization control in order to execute a plurality of OSs with one information processing apparatus.

  Reference numerals 601 and 602 denote OSs (an example of an intra-cycle processing unit) executed on the hypervisor. Two or more OSs may exist. Further, instead of the OS, a virtual machine (also referred to as VM) operating on the hypervisor may be used. In the case of a virtual machine (an example of an in-cycle processing unit), a virtual machine monitor may be used instead of the hypervisor 501.

  It is assumed that the hypervisor 501 has a CPU scheduling function unit 111 that performs upper limit management of CPU allocation time for the OS or virtual machine. Since the scheduling performed by the hypervisor for the OS or the virtual machine is the same as the scheduling performed by the OS for the task, the free resource management function unit 121 described in the first embodiment is assigned to the CPU scheduling of the hypervisor 501. Can be operated in combination with functions. The configuration in the free resource management function unit 121 is the same as that of the first embodiment as shown in FIG. The operations of the CPU scheduling function unit 111 and the free resource management function unit 121 are the same as those shown in the flowcharts of FIGS. 2, 5, and 6.

(Reading OS etc.)
In the case of the third embodiment, the OS 101 that performs CPU scheduling is replaced with the hypervisor 501, and the task that is subject to CPU scheduling is replaced with the OS (or virtual machine). By this replacement, the CPU allocation time management method described in the first embodiment can be applied to the scheduling of the hypervisor 501. The difference between FIG. 8 corresponding to this replacement and FIG. 1 of the first embodiment is surrounded by a broken line.

  As described above, in the third embodiment, the CPU allocation time management method described in the first embodiment is incorporated in the CPU allocation time management of a hypervisor that executes a plurality of OSs (or VMs) with one information processing apparatus. Thus, CPU allocation control can be performed when CPU free time occurs in the OS (or VM). Thus, the CPU free time allocation method can be changed in accordance with whether the OS executed on the hypervisor 501 requires high speed or constant performance.

Embodiment 4 FIG.
The fourth embodiment will be described with reference to FIGS. In the fourth embodiment, a hardware configuration of a CPU allocation time management apparatus 1000 that is a computer will be described. FIG. 9 is a diagram showing an example of the appearance of a CPU allocation time management apparatus 1000 that is a computer. FIG. 10 is a diagram illustrating an example of hardware resources of the CPU allocation time management apparatus 1000 described in the first to third embodiments.

  In FIG. 9 showing an appearance, a CPU allocation time management apparatus 1000 includes a system unit 830, a display device 813 having a CRT (Cathode / Ray / Tube) or LCD (liquid crystal) display screen, a keyboard 814 (Key / Board: K / B), hardware resources such as a mouse 815, a compact disk device 818 (CDD: Compact Disk Drive), and the like, which are connected by cables and signal lines. The system unit 830 is connected to the network.

  In FIG. 10 showing hardware resources, the CPU allocation time management apparatus 1000 includes a CPU 810 (Central Processing Unit) for executing a program. The CPU 810 is connected to a ROM (Read Only Memory) 811, a RAM (Random Access Memory) 812, a display device 813, a keyboard 814, a mouse 815, a communication board 816, a CDD 818, and a magnetic disk device 820 via a bus 825. Control hardware devices. Instead of the magnetic disk device 820, a storage device such as an optical disk device or a flash memory may be used.

  The RAM 812 is an example of a volatile memory. Storage media such as the ROM 811, the CDD 818, and the magnetic disk device 820 are examples of nonvolatile memories. These are examples of a “storage device” or a storage unit, a storage unit, and a buffer. These storage devices and the like realize a “queue”. The communication board 816, the keyboard 814, and the like are examples of an input unit and an input device. The communication board 816, the display device 813, and the like are examples of an output unit and an output device. The communication board 816 is connected to the network.

  The magnetic disk device 820 stores an operating system 821 (OS), a window system 822, a program group 823, and a file group 824. The programs in the program group 823 are executed by the CPU 810, the operating system 821, and the window system 822.

  The OS 821 and the program group 823 store programs that execute the functions described as “˜units” in the description of the above embodiments. The program is read and executed by the CPU 810.

  In the description of the above embodiment, the file group 824 includes “to determination result”, “to calculation result”, “to extraction result”, “to generation result”, and “to processing result”. The described information, data, signal values, variable values, parameters, and the like are stored as items of “˜file” and “˜database”. The “˜file” and “˜database” are stored in a recording medium such as a disk or a memory. Information, data, signal values, variable values, and parameters stored in a storage medium such as a disk or memory are read out to the main memory or cache memory by the CPU 810 via a read / write circuit, and extracted, searched, referenced, compared, and calculated. Used for CPU operations such as calculation, processing, output, printing, and display. Information, data, signal values, variable values, and parameters are temporarily stored in the main memory, cache memory, and buffer memory during the CPU operations of extraction, search, reference, comparison, operation, calculation, processing, output, printing, and display. Is remembered.

  In the description of the embodiment described above, the data and signal values are the memory of the RAM 812, the compact disk of the CDD 818, the magnetic disk of the magnetic disk device 820, other optical disks, mini disks, and DVDs (Digital Versatile Disk). Or the like. Data and signals are transmitted on-line via the bus 825, signal lines, cables, and other transmission media.

  In the above description of the embodiment, what has been described as “to part” may be “to means”, and “to step”, “to procedure”, and “to processing”. May be. That is, what has been described as “˜unit” may be implemented by software alone, a combination of software and hardware, or a combination of firmware. Firmware and software are stored as programs in a recording medium such as a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, and a DVD. The program is read by the CPU 810 and executed by the CPU 810. That is, the program causes the computer to function as the “˜unit” described above. Alternatively, the computer executes the procedure and method of “to part” described above.

  Although the CPU allocation time management apparatus 1000 has been described in the above embodiment, it is obvious from the above description that the operation of the CPU allocation time management apparatus 1000 can be grasped as a program for causing a computer to execute the operation. It is also possible to grasp the operation of the CPU allocation time management apparatus 1000 described in the above embodiment as CPU allocation time management performed by each unit of the CPU allocation time management apparatus 1000.

(1) According to the CPU allocation time management apparatus 1000 of the first embodiment described above, the upper limit control of the CPU allocation time is performed for each task group. A CPU allocation time system that can be controlled can be provided.
(2) Further, according to the CPU allotted time management apparatus 1000 of the second embodiment, it is possible to obtain a method capable of dynamically changing parameters necessary for such management when the information processing apparatus is executed.
(3) In addition, according to the CPU allocation time management apparatus 1000 of the third embodiment, such a CPU allocation time management method is only used for CPU allocation time control of a task group composed of tasks in one OS. Instead, it can be used for CPU allocation time control between OSs (or between VMs) on a hypervisor that provides a function of simultaneously executing a plurality of OSs (or a plurality of VMs).

In the above embodiment,
An OS of an information processing apparatus that executes a plurality of tasks simultaneously (strictly, time-division multitasking). For a task group composed of one or more task groups, the CPU time that can be used by the task group In the CPU allocation time management system for setting the upper limit, the CPU allocation time management system including the following means has been described.
(A) A free resource pool for reserving CPU free time when CPU free time occurs;
(B) Resource restriction mode information for selecting whether or not to perform CPU time allocation exceeding the upper limit of the CPU time allocated to the task group when CPU idle time occurs.
(C) An allocation overtask queue in which tasks to be executed exceeding the upper limit of CPU time allocated to a task group when a CPU idle time occurs are registered.
(D) Called when a free time occurs in the CPU time allocated to the task group, reserves the CPU free time by the free resource pool, and determines whether to allocate the CPU free time to the task group based on the resource limit mode information. Free resource management function to control and execute tasks registered in the assigned overtask queue.

In the above embodiment, the CPU allocation time management system characterized by including the following means has been described.
(A) CPU time management cycle setting for clearing CPU usage time information to be used for CPU allocation time management and allocation overtask queue information and resetting the state of the free resource management function for each specified cycle time.

In the above embodiment, the CPU allocation time management system characterized by including the following means has been described.
(A) Free resource available group setting for assigning CPU free time only to a specified task group when CPU free time occurs and allocation of CPU free time is permitted by resource restriction mode information .

In the above embodiment, the CPU allocation time management system characterized by including the following means has been described.
(A) A free resource management parameter setting function for changing setting values registered in the resource restriction mode, CPU time management cycle setting, and free resource available group setting during OS execution.

In the above embodiment, the following CPU allocation time management system has been described.
A CPU used by the OS or virtual machine for the OS or virtual machine to be executed by the hypervisor or virtual machine monitor of the information processing apparatus for executing a plurality of OSs or virtual machines simultaneously (time division multitasking) In the CPU allocation time management system for setting an upper limit of time, the CPU allocation time management system having the CPU allocation time management system described above in a hypervisor or a virtual machine monitor.

  1000 CPU allocation time management apparatus, 101 OS, 111 CPU scheduling function unit, 112 CPU allocation time table, 113 scheduling queue, 121 free resource management function unit, 122 free resource pool, 123 resource restriction mode, 124 free resource available group setting , 125 CPU time management cycle setting, 126 allocation overtask queue, 131 free resource management parameter setting function unit, 201 group 1, 202 group 2, 501 hypervisor, 601 and 602 OS (or VM).

Claims (10)

CPU allocated time indicating the upper limit value of the continuous use time of the CPU, which is a processing unit of the CPU within the execution cycle scheduled for each execution cycle set for the CPU (Central Processing Unit). A first execution unit that causes a CPU to execute an intra-period processing unit that is a processing unit for which the CPU is set within a CPU allocation time;
When a plurality of in-cycle processing units are executed by the CPU in any execution cycle, an in-cycle unfinished unit that is an in-cycle processing unit in which execution by the CPU does not end occurs during the CPU allocation time, An in-cycle processing unit that is executed after an in-cycle unfinished unit during the execution cycle in which the in-cycle unfinished unit is generated, and that is executed within a period shorter than the CPU allocation time. A CPU allocation time, comprising: a second execution unit that causes a CPU to execute an unfinished unit within a cycle following the end of execution of the early termination unit within the cycle when an early termination unit within the cycle occurs. Management device. The second execution unit is
When the hardware restriction mode setting instructing not to execute the unfinished unit within the cycle is accepted, and when the hardware restricted mode setting is accepted, the CPU does not execute the unfinished unit within the cycle even if the early termination unit within the cycle exists. The CPU allocated time management device according to claim 1. The second execution unit is
The specific processing unit within a cycle is accepted as an unfinished execution target, and the CPU is caused to execute only when the intra-cycle early termination unit exists only for the intra-cycle processing unit accepted as an unfinished execution target. 3. The CPU allocated time management device according to any one of 2 above. The CPU allotted time management device further includes:
An execution cycle setting unit for setting the execution cycle of the CPU;
The CPU allocated time management apparatus according to claim 1, further comprising a cycle changing unit that changes the execution cycle set by the execution cycle setting unit. The CPU allotted time management device further includes:
The second execution unit is instructed to switch between a software restriction mode setting for instructing execution by the CPU of a non-ending unit within a cycle and a hardware restriction mode setting when there is an early termination unit within the cycle. A mode switching unit to
The second execution unit is
3. The CPU allocated time management apparatus according to claim 2, wherein either the soft limit mode setting or the hard limit mode setting is executed in response to a switching instruction from the mode switching unit. The CPU allotted time management device further includes:
The CPU allocated time management device according to claim 3, further comprising an unfinished execution target changing unit that changes an unfinished execution target received by the second execution unit. The in-cycle processing unit is
The CPU allocation time management apparatus according to claim 1, wherein the CPU allocation time management apparatus is a task group including one or more tasks. The in-cycle processing unit is
7. The CPU allocation time management apparatus according to claim 1, wherein the CPU allocation time management apparatus is an OS (Operating System). The in-cycle processing unit is
The CPU allocation time management apparatus according to claim 1, wherein the CPU allocation time management apparatus is a virtual machine. Computer
CPU allocated time indicating the upper limit value of the continuous use time of the CPU, which is a processing unit of the CPU within the execution cycle scheduled for each execution cycle set for the CPU (Central Processing Unit). A first execution unit that causes the CPU to execute the intra-period processing unit that is a processing unit for which the CPU is set within the CPU allotted time;
When a plurality of in-cycle processing units are executed by the CPU in any execution cycle, an in-cycle unfinished unit that is an in-cycle processing unit in which execution by the CPU does not end occurs during the CPU allocation time, An in-cycle processing unit that is executed after an in-cycle unfinished unit during the execution cycle in which the in-cycle unfinished unit is generated, and that is executed within a period shorter than the CPU allocation time. A second execution unit that causes the CPU to execute the non-terminating unit within the cycle following the end of the execution of the early term termination unit within the cycle when the early termination unit within the cycle occurs;
CPU allocation time management program for functioning as

Images