Information processing apparatus and method, vehicle, and computer-readable storage medium

Abstract

The present disclosure provides an information processing apparatus, a vehicle, an information processing method, and a computer-readable storage medium. In a startup phase, in order to quickly start the overall management VM0 having a function required for initialization of each VM to work, the CPU allocation time is set to 1000 μs for VM0 and 0 μs for VM1 and VM2, in a normal phase, the CPU allocation time is set to 200 μs for VM0 and 400 μs for each of VM1 and VM2, in a sleep phase, the CPU allocation time is set to 700 μs for VM0 and 150 μs for each of VM1 and VM2, and in a special processing phase, the CPU allocation time to the VM2 performing special processing is increased. Furthermore, the setting section forcibly shifts to the normal phase by setting the CPU allocation time in the manner of the normal phase and shifting to the normal phase in the case where there is no phase shift request within a predetermined time after the switching to the phase is performed.

Description

Technical Field

[0001] This disclosure relates to an information processing apparatus, a vehicle, an information processing method, and a computer-readable storage medium storing an information processing program that can be applied to a vehicle system. Background Technology

[0002] Japanese Patent Application Publication No. 2016-091109 (Patent Document 1) discloses a dynamic resource allocation device that calculates the amount of resources allocated to a virtual machine and dynamically allocates resources. Specifically, this resource allocation device includes: a usage calculation unit that calculates a determined usage amount, which is the actual usage amount of resources in each time slot divided by the changing cycle of resources implemented by the virtual machine; a peak detection unit that detects peaks in the determined usage amount; an allocation amount calculation unit that calculates the allocation amount of resources allocated to the i-th time slot based on the past determined usage amount in the i-th time slot and the detection results of past peaks in time slots within a predetermined range before and after the i-th time slot; and an allocation amount setting unit that sets the allocation amount to a virtual machine monitor that controls the virtual machine. Summary of the Invention

[0003] The problem that the invention aims to solve

[0004] In systems like in-vehicle systems where real-time performance is essential, there is room for improvement because dynamically adjusting resource allocation timing makes system validity verification difficult.

[0005] This disclosure is an invention made in consideration of the foregoing facts, and its object is to provide an information processing apparatus, vehicle, information processing method, and computer-readable storage medium storing an information processing program that can be applied to a system where resource allocation time is adjustable and real-time performance is required.

[0006] Methods for solving problems

[0007] The information processing apparatus according to the first embodiment includes: a generation unit that generates multiple virtual machines, including a management virtual machine, which manages the multiple virtual machines; a detection unit that detects multiple predetermined stages; and a setting unit that, based on the detection results of the detection unit, sets the resource allocation time of the multiple virtual machines in a manner that forms a predetermined scheduling table for each stage, and sets the allocation time in a manner that forms a predetermined basic stage scheduling table for allocating the allocation time to each of the multiple virtual machines if no stage switching request is made within a predetermined time after a switch to a predetermined stage is implemented.

[0008] According to the first embodiment, in the generation unit, multiple virtual machines, including a management virtual machine, are generated, and the management virtual machine manages the multiple virtual machines.

[0009] In the detection unit, multiple pre-defined stages are tested. In the setting unit, based on the test results, the resource allocation time for multiple virtual machines is set in a pre-defined scheduling table for each stage. This allows for changes to the resource allocation time for each stage. Furthermore, since the CPU allocation time is fixed and statically scheduled in each stage, the necessary real-time performance can be guaranteed in the vehicle system. Therefore, an information processing device applicable to systems where resource allocation time is adjustable and real-time performance is required can be provided.

[0010] Furthermore, in the configuration section, if no stage switching request is made within a predetermined time after a switch to a pre-defined stage has been performed, the allocation time is set as a schedule table of pre-defined basic stages that allocates the allocation time to each of the multiple virtual machines. Therefore, even if the managed virtual machine malfunctions for some reason and the stage is not switched, the allocation time is still set as a schedule table of basic stages, thus preventing problems such as system startup failures.

[0011] Alternatively, as a second implementation, the plurality of stages may include a normal stage as the basic stage, wherein the normal stage is a stage in which the allocation time for the plurality of virtual machines is set to a predetermined normal time. This allows for the allocation of time required for normal operation to each virtual machine and makes it possible to handle anomalies.

[0012] Furthermore, as a third implementation, the plurality of stages may include a startup stage, which is a stage in which the allocation time is set only for the management virtual machine, or the allocation time of the management virtual machine is set to a time that is longer than the allocation times of the other virtual machines. The pre-defined time includes the startup stage time specified corresponding to the startup stage. This allows for the rapid startup of the management virtual machine, which has the initialization and other functions required for each virtual machine to operate.

[0013] Furthermore, as a fourth embodiment, the plurality of stages may include a sleep stage, which is a stage in which the allocation time of the managed virtual machine is set to be longer than the allocation time of other virtual machines, and the predetermined time includes a sleep stage time specified corresponding to the sleep stage. This shortens the recovery time from sleep.

[0014] Furthermore, as a fifth embodiment, the plurality of stages may include a special processing stage, which is a stage in which the allocation time is set only for special virtual machines that perform pre-defined special processing, or the allocation time of the special virtual machine is set to be longer than the allocation time of other virtual machines. The pre-defined time includes a special processing stage time specified corresponding to the special processing stage. This allows resources to be concentrated on the special virtual machines that perform special processing.

[0015] Alternatively, as a sixth embodiment, it can be a vehicle that includes the information processing device described in any one of the first to fifth embodiments.

[0016] Furthermore, as a seventh embodiment, it can also be configured as an information processing method, wherein the information processing method is performed by a computer as follows: generating multiple virtual machines, including a management virtual machine, which manages the multiple virtual machines; detecting multiple predefined stages; setting the resource allocation time of the multiple virtual machines in a manner that is a predefined scheduling table for each stage based on the detection results of the stages; and setting the allocation time in a manner that is a predefined basic stage scheduling table for allocating the allocation time to each of the multiple virtual machines if no stage switching request is made within a predefined time after switching to a predefined stage.

[0017] Alternatively, as an eighth embodiment, a computer-readable storage medium may be provided, storing an information processing program for causing a computer to perform the following processes, wherein the processes include: generating multiple virtual machines, including a management virtual machine, which manages the multiple virtual machines; detecting multiple predefined stages; setting the resource allocation time of the multiple virtual machines in a manner that is a predefined scheduling table for each stage based on the detection results of the stages; and setting the allocation time in a manner that is a predefined scheduling table for allocating the allocation time to each of the multiple virtual machines if no stage switching request is made within a predefined time after a switch to a predefined stage has been implemented.

[0018] Invention Effects

[0019] As explained above, according to this disclosure, it is possible to provide an information processing apparatus, vehicle, information processing method, and computer-readable storage medium storing an information processing program that can be applied to a system where resource allocation time is adjustable and real-time operation is required. Attached Figure Description

[0020] Figure 1 A diagram showing a vehicle equipped with the central ECU according to this embodiment.

[0021] Figure 2 This is a block diagram illustrating the general structure of the central ECU involved in this embodiment.

[0022] Figure 3 A function block diagram representing the functions of the hypervisor.

[0023] Figure 4 A diagram illustrating an example of scheduling for CPU time allocation at each stage.

[0024] Figure 5 This is a diagram used to illustrate the transition between stages.

[0025] Figure 6 This is a flowchart illustrating an example of the processing flow when setting the CPU allocation time implemented by the central ECU according to this embodiment. Detailed Implementation

[0026] Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In this embodiment, a central ECU mounted in a vehicle will be described as an example of an information processing device. In this embodiment, Figure 1 The diagram shows a vehicle equipped with the central ECU (Electronic Control Unit) according to this embodiment. Figure 2 This is a block diagram illustrating the general structure of the central ECU involved in this embodiment.

[0027] The central ECU 12 involved in this embodiment is mounted on the vehicle 10 and performs comprehensive control over the various ECUs installed on the vehicle 10.

[0028] The central ECU12 is equipped with a CPU (Central Processing Unit). In this embodiment, for example... Figure 2 As shown, multiple CPU cores exist as an example (in...) Figure 2 In the example, these are CPU cores 1 through 4) 14.

[0029] In this embodiment, the physical CPU core 14 is virtualized by a hypervisor 16, which acts as software for virtualizing the computer, thereby generating VMs (Virtual Machines) 18. In this embodiment, multiple VMs 18 are generated by the hypervisor 16. Figure 2 The example shown illustrates how three VM18s, VM0 to VM2, were generated as multiple VM18s.

[0030] An OS (Operating System) 20 is configured on each VM18, and applications (Apps) 22 operate on OS 20. Figure 2 In this system, Apps 1 and 2 operate on OS 1, Apps 3 and 4 operate on OS 2, and Apps 5 and 6 operate on OS 3.

[0031] Furthermore, in typical management programs, the function of managing each VM is integrated into the management program itself. However, as in this embodiment, the management program 16 used in the vehicle system has its functionality greatly reduced to ensure real-time performance, and the function of managing each VM 18 is configured as a single VM 18. In this embodiment, VM0 functions as a management virtual machine that manages each VM 18; in the following text, VM0 may be referred to as the overall management VM 18. Moreover, setting up an overall management VM 18 increases the integration between the overall management VM 18 and each VM 18, thus creating a dependency relationship.

[0032] Furthermore, since multiple VMs 18 are configured in the hypervisor 16, by allocating CPU time to each VM 18, it is possible for each VM 18 to appear to be working in parallel.

[0033] In order for each VM18 to work effectively, it is desirable to dynamically change the scheduling of CPU time allocation as a resource. However, in automotive systems, dynamically changing the scheduling makes it difficult to guarantee real-time performance.

[0034] Therefore, in the central ECU 12 of this embodiment, scheduling is performed semi-dynamically by changing the CPU time allocation schedule according to each stage. This allows for the application of appropriate scheduling for each stage. Furthermore, since the scheduling would be static if viewed on a stage-by-stage basis, real-time performance is guaranteed.

[0035] Here, the functional structure of the scheduling management program 16, which is used to change the CPU allocation time according to each stage, will be explained. Figure 3A function block diagram to represent the functions of management program 16.

[0036] like Figure 3 As shown, the management program 16 has the functions of a generation unit 24, a detection unit 26, and a setting unit 28.

[0037] The generation unit 24 generates and executes multiple VMs 18 formed by virtualizing the physical CPU core 14. In this embodiment, three VMs, VM0 to VM2, are generated as described above.

[0038] The detection unit 26 detects multiple stages used to change CPU allocation time. In this embodiment, it detects four stages: startup stage, normal stage, sleep stage, and special processing stage.

[0039] Based on the detection results from the detection unit 26, the setting unit 28 modifies the CPU allocation time in a manner that forms a pre-defined schedule for each stage, thereby setting the CPU allocation time schedule. In this embodiment, the CPU allocation time schedule is modified according to each of the startup stage, normal stage, sleep stage, and special processing stage.

[0040] Figure 4 This is a diagram illustrating an example of scheduling where CPU time is allocated at each stage. Figure 4 The diagram illustrates an example of CPU time allocation scheduling for the startup phase, normal phase, sleep phase, and special processing phase. In each phase, the Major Time Frame is set to 1000 μs.

[0041] During the startup phase, the overall management VM18 (VM18) is initialized to quickly start up the functions required for each VM18 to work. Figure 2 VM0), thus allowing VM0 to occupy CPU time. That is, in Figure 4 During the startup phase, VM0 is set to 1000μs, VM1 and VM2 are set to 0μs, and after initializing the overall management VM18, it transitions to the normal phase. Additionally, although in Figure 4 The example shown is of VM0 occupying CPU time, but it is not limited to this. The CPU allocation time of VM0 can also be set to be longer compared with the CPU allocation time of other VM1 and VM2.

[0042] During the normal phase, the CPU allocation time required for normal operation is set for each VM18 as a pre-defined normal time. Figure 4 The example shows an instance where 200 μs is allocated to VM0, and 400 μs is allocated to VM1 and VM2 respectively.

[0043] During the sleep phase, because the data saving requests for VM0, which is managed as non-volatile memory, are concentrated, the CPU time allocated to the overall management VM18 is increased, and the sleep time is shortened. Figure 4 The example shows an instance where 700 μs is allocated to VM0, and 150 μs is allocated to VM1 and VM2 respectively.

[0044] During special processing phases, such as when special processing is performed on VM2 as a special virtual machine, the CPU time allocated to VM2 performing special processing is increased, and special processing is prioritized for execution. Figure 4 The example shows an instance where 200 μs are allocated to VM0 and VM1 respectively, and 600 μs is allocated to VM2. Furthermore, regarding the special processing phase, an example of implementing collision detection and other processing will be given below. Moreover, since special processing such as collision detection is implemented during the special processing phase, it is also possible to allocate CPU time only to VM2, thus allowing VM2 to occupy CPU time.

[0045] Thus, by changing the scheduling of CPU allocation time to each VM18 according to each stage to adjust the resource allocation time, the CPU allocation time is fixed and statically scheduled in each stage, thereby ensuring the necessary real-time performance in the vehicle system.

[0046] However, if, for some reason, the overall management VM18 or similar system malfunctions, resulting in the failure to receive an initialization completion notification during the startup phase and thus preventing phase switching, all systems will become unable to boot. Furthermore, in other phases, without receiving a phase switching delegation, it will also become impossible to return to the normal phase.

[0047] Therefore, in this embodiment, if no stage switching request is made within a predetermined time after a stage switch has been performed, the setting unit 28 sets the allocation time to a normal stage, which is a pre-defined basic stage in the scheduling table, thereby forcibly transferring to a normal stage. Furthermore, although an example of applying a normal stage scheduling table as a basic stage scheduling table is described below, it is not limited to this. For example, a basic stage may be further set in addition to a normal stage, and the CPU allocation time pre-set as a basic stage may be set as the basic stage scheduling table. Alternatively, the CPU allocation time of other stages may be set as the basic stage scheduling table.

[0048] Specifically, such as Figure 5As shown, if the setting unit transfers to each stage and there is still no stage switching request after a pre-defined time for each stage, it sets the allocation time to become a normal stage, thereby switching to a normal stage.

[0049] exist Figure 5 In the example, even after the transition to the startup phase at startup, which is a pre-defined time, the first time (used for the startup phase) has elapsed (in the example). Figure 5 If, for example, 2000ms has elapsed but no initialization completion notification has been received, the system switches to the normal phase.

[0050] Furthermore, through sleep transition notification, even if a second time (the time allotted for the sleep stage) has elapsed after the transition from the normal stage to the sleep stage (as per a pre-defined timeframe),... Figure 5 If, as an example, the sleep process has not yet been initiated (e.g., 5000ms), the system switches to the normal phase. When transitioning from the sleep phase to the normal phase, the system switches via the initiation phase.

[0051] Furthermore, through collision detection notification, even after a pre-defined time has elapsed since the transition from the normal phase to the special handling phase (in the third time period designated for the special handling phase), the collision detection notification ensures that the time period is sufficient for the special handling phase. Figure 5 If, as an example, the time is 100ms but no notification is received that the post-collision processing has ended, the system switches to the normal phase.

[0052] in addition, Figure 5 The pre-defined time for reverting from each stage to the normal stage shown is an example and is not limited to the time mentioned above.

[0053] Next, the specific processing implemented by the central ECU12 of this embodiment, which is configured as described above, will be explained. Figure 6 This is a flowchart illustrating an example of the processing flow for setting the CPU allocation time implemented by the central ECU 12 according to this embodiment. Furthermore, Figure 6 The process begins, for example, when the vehicle's power is turned on, such as by an ignition switch (not shown).

[0054] In step 100, the management program 16 is configured to allocate CPU time for the startup phase and proceeds to step 102. That is, when the vehicle's power is turned on, the detection unit 26 detects the startup phase, and the setting unit 28 configures the CPU allocation time for the startup phase. Specifically, as follows... Figure 4The settings are set as follows: 1000 μs is set for VM0, which manages VM18 as a whole, and 0 μs is set for VM1 and VM2, so that VM0, which manages VM18 as a whole, occupies CPU time.

[0055] In step 102, the management program 16 determines whether it is transitioning to the normal phase or whether there is no phase switching request within the first time period of the startup phase after the phase switch was implemented. This determination, for example, involves checking whether the detection unit 26 received a startup phase end notification from the overall management VM 18 or whether there is no phase switching request within 2000ms after switching to the startup phase. The program then waits until this determination is affirmative and proceeds to step 104. Alternatively, if in step 102, the determination is affirmative because there is no phase switching request within the first time period of the startup phase, the setting unit 28 forcibly switches the phase to the normal phase during the transition to step 104.

[0056] In step 104, the management program 16 is changed to allocate CPU time for the normal phase, and the process proceeds to step 106. That is, the setting unit 28 is changed to set the CPU time allocation for the normal phase. Specifically, as follows... Figure 4 As shown, the CPU allocation time is changed so that VM0 is 200μs and VM1 and VM2 are 400μs respectively.

[0057] In step 106, the management program 16 determines whether a transition to the sleep stage has occurred. This determination, for example, involves checking whether the detection unit 26 has detected the fulfillment of pre-defined conditions for the transition to the sleep stage. If the determination is affirmative, the program proceeds to step 108; otherwise, it proceeds to step 112.

[0058] In step 108, the management program 16 is changed to set the CPU allocation time for the sleep stage, and the process proceeds to step 110. That is, the setting unit 28 is changed to set the CPU allocation time for the sleep stage. Specifically, as follows... Figure 4 As shown, 700μs is set for VM0, which manages VM18 as a whole, and 150μs is set for VM1 and VM2 respectively, thereby increasing the CPU allocation time for managing VM18 as a whole and shortening the time for transitioning to sleep and resuming.

[0059] In step 110, the management program 16 determines whether the system has resumed from sleep or whether there is no stage switching request within a second time period (the sleep stage time) after a stage switch has been performed. This determination, for example, involves the detection unit 26 detecting whether a pre-defined condition for resuming from sleep has been met, or whether there is no stage switching request within 5000ms after switching to the sleep stage. If the determination is positive, the system returns to step 100 and repeats the above process; otherwise, it proceeds to step 112. Furthermore, although this embodiment shows an example of returning to step 100 and setting the system to the startup stage after resuming from the sleep stage, it is not limited to this; it is also possible to transition to the normal stage after resuming from the sleep stage. In this case, if the determination in step 110 is positive, the system proceeds to step 104. Additionally, if in step 110, the determination is positive because there is no stage switching request within the second time period (the sleep stage time), the setting unit 28 forcibly switches the stage to the normal stage via the startup stage when transitioning to step 100.

[0060] In step 112, the management procedure 16 determines whether a transition to a special processing stage has occurred. This determination, for example, involves checking whether the detection unit 26 has detected the fulfillment of pre-defined conditions for transitioning to the special processing stage. If the determination is affirmative, the process proceeds to step 114; otherwise, it proceeds to step 118.

[0061] In step 114, the management program 16 is changed to allocate CPU time for the special processing stage, and the process proceeds to step 116. That is, the setting unit 28 is changed to set the CPU time allocation for the special processing stage. Specifically, as follows... Figure 4 As shown, the CPU allocation time is changed so that VM0 and VM1 are each 200μs and VM2 is 600μs.

[0062] In step 116, the management program 16 determines whether the process has resumed from a special processing phase or whether there is no phase switching request within the third time period of the special processing phase after the phase switch was implemented. This determination, for example, involves checking whether the detection unit 26 detects the fulfillment of a pre-defined condition for resuming from the special processing phase or whether there is no phase switching request within 100ms after switching to the sleep phase. If the determination is positive, the process returns to step 104 and repeats the above process; otherwise, it proceeds to step 118. Furthermore, if the determination is positive because there is no phase switching request within the third time period of the special processing phase, the setting unit 28 forcibly switches the phase to the normal phase when transitioning to step 104.

[0063] In step 118, the management program 16 determines whether to end the process. This determination may include, for example, checking whether the detection unit 26 has detected that the vehicle's power supply, such as an ignition switch (not shown), has been disconnected. If the determination is negative, the process proceeds to step 120; otherwise, if the determination is positive, the entire process ends.

[0064] In step 120, the management procedure 16 determines whether a special processing stage is in progress. This determination is made by confirming whether a special processing stage is in progress, and if so, a positive determination is made, returning to step 116 and repeating the aforementioned processing. Conversely, if a sleep stage or a normal stage is in progress, a negative determination is made, and the process proceeds to step 122.

[0065] In step 122, the management program 16 determines whether it is in a normal phase. This determination is made by checking if it is in a normal phase, and if so, a positive determination is made and the process returns to step 106, repeating the above process. Conversely, if it is in a sleep phase, a negative determination is made and the process returns to step 110, repeating the above process.

[0066] By processing in this way, it is possible to change the scheduling of CPU time allocation to each VM18 according to each stage, and to adjust the resource allocation time.

[0067] Furthermore, since the CPU allocation time is fixed and statically scheduled in each stage, the necessary real-time performance can be guaranteed in the vehicle system.

[0068] In addition, even if the overall management VM18 fails for some reason and the phase is not switched, it will be forcibly switched to the normal phase and set to allocate CPU time for the normal phase. Therefore, it can suppress the occurrence of problems such as system failure to start.

[0069] Furthermore, although the above embodiment describes an example with four CPU cores 14, it is not limited to this. For example, it can be configured to have a single CPU core 14, or it can be configured to have multiple CPU cores other than four.

[0070] Furthermore, although the example of the hypervisor 16 generating three VMs 18 was described in the above embodiment, it is not limited to this. For example, it can generate two VMs 18, or it can generate more than four VMs 18.

[0071] Furthermore, although the above embodiments have been described using four stages—start-up stage, normal stage, sleep stage, and special processing stage—as an example, the stages are not limited to these four stages. For example, there may be multiple different stages other than these four stages, two or three of the above four stages, or multiple stages that add other stages to the above four stages.

[0072] Furthermore, the processing performed by the management program 16 in the above embodiments can also be stored as a program in various storage media for distribution.

[0073] Furthermore, this disclosure is not limited to the above content, and various modifications can be made without departing from its main idea.

Claims

1. An information processing apparatus, comprising: The generation department generates multiple virtual machines, including a management virtual machine, which manages the multiple virtual machines. The testing department conducts testing at multiple pre-defined stages; The setting unit, based on the detection results of the detection unit, sets the resource allocation time of the plurality of virtual machines in a manner that is a pre-defined scheduling table for each of the aforementioned stages. Furthermore, if no stage switching request is made within a pre-defined time after a switch to a pre-defined stage has been implemented, the allocation time is forcibly set in a manner that is a pre-defined scheduling table of a basic stage that allocates the allocation time to each of the plurality of virtual machines.

2. The information processing apparatus as described in claim 1, wherein, The plurality of stages includes a normal stage as the basic stage, which is a stage in which the allocation time of the plurality of virtual machines is set to a pre-defined normal time.

3. The information processing apparatus as described in claim 1 or claim 2, wherein, The plurality of phases includes a startup phase, which is a phase in which the allocation time is set only for the management virtual machine, or the allocation time of the management virtual machine is set to be longer than the allocation time of other virtual machines, and the pre-defined time includes the startup phase time defined corresponding to the startup phase.

4. The information processing apparatus according to any one of claims 1 to 3, wherein, The plurality of phases includes a sleep phase, which is a phase in which the allocation time of the managed virtual machine is set to be longer than the allocation time of other virtual machines, and the predefined time includes a sleep phase time defined corresponding to the sleep phase.

5. The information processing apparatus according to any one of claims 1 to 4, wherein, The plurality of stages includes a special processing stage, which is a stage in which the allocation time is set only for a special virtual machine that performs the pre-defined special processing, or the allocation time of the special virtual machine is set to be longer than the allocation time of other virtual machines, and the pre-defined time includes the special processing stage time specified corresponding to the special processing stage.

6. A vehicle, wherein, The device comprises the information processing apparatus according to any one of claims 1 to 5.

7. An information processing method, wherein, The computer performs the following processing: Multiple virtual machines are generated, including a management virtual machine, which manages the multiple virtual machines. Testing is conducted at multiple pre-defined stages; Based on the detection results of the aforementioned stages, the resource allocation time for the plurality of virtual machines is set in a manner that is a pre-defined scheduling table for each of the aforementioned stages. Furthermore, if no stage switching delegation is requested within a pre-defined time after a switch to a pre-defined stage has been implemented, the allocation time is forcibly set in a manner that is a pre-defined scheduling table for a basic stage that allocates the allocation time to each of the plurality of virtual machines.

8. A computer-readable storage medium, which is a non-transitory computer-readable storage medium storing an information processing program for causing a computer to perform the following processes, wherein, The process includes: Multiple virtual machines are generated, including a management virtual machine, which manages the multiple virtual machines. Testing is conducted at multiple pre-defined stages; Based on the detection results of the aforementioned stages, the resource allocation time for the multiple virtual machines is set in a manner that is a pre-defined scheduling table for each stage. Furthermore, if no stage switching delegation is requested within a pre-defined time after a switch to a pre-defined stage has been implemented, the allocation time is forcibly set in a manner that is a pre-defined scheduling table for a basic stage that allocates the allocation time to each of the multiple virtual machines.

Images