Information processing device, vehicle, information processing method, and information processing program

By employing semi-dynamic scheduling for virtual machines in in-vehicle systems, the challenge of maintaining real-time performance during resource allocation is addressed through phase-specific CPU time adjustments, ensuring efficient and reliable system operation.

JP7586052B2Active Publication Date: 2024-11-19TOYOTA JIDOSHA KK
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2021180653
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-11-04
Publication Date
2024-11-19
Estimated Expiration
2041-11-04

AI Technical Summary

Technical Problem

In systems requiring real-time performance, such as in-vehicle systems, dynamically adjusting resource allocation time makes it difficult to verify system feasibility.

Method used

A semi-dynamic scheduling approach is implemented by generating multiple virtual machines, including a management virtual machine, to detect and adjust resource allocation times for different phases, ensuring static scheduling within each phase to guarantee real-time performance.

Benefits of technology

This approach allows for efficient resource allocation adjustment while maintaining real-time performance by fixing CPU allocation times within each phase, thereby ensuring reliable operation of in-vehicle systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007586052000001
    Figure 0007586052000001
  • Figure 0007586052000002
    Figure 0007586052000002
  • Figure 0007586052000003
    Figure 0007586052000003
Patent Text Reader

Abstract

To provide an information processing apparatus, a vehicle, an information processing method, and an information processing program which can be applied to a system in which allocation time of resources is adjustable and real-time property is required.SOLUTION: In a start-up phase, an overall management virtual machine (VM) VM0 having initialization and a function necessary for each VM in order to operate is started quickly, CPU allocation time is assume that a VM0 has of 1000 μs, a VM1 and the a VM2 have 0 μs, and in a normal phase, the VM0 has 200 μs, the VM1 and the VM2 have each 400 μs respectively, and in a sleep phase, the VM0 is 700 μs, and the VM1 and the VM 2 have 150 μs respectively.SELECTED DRAWING: Figure 4
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an information processing device, a vehicle, an information processing method, and an information processing program that can be applied to a vehicle system. [Background technology]

[0002] Patent Document 1 proposes a dynamic resource allocation device that calculates a resource allocation amount to a virtual machine and dynamically allocates the resource. In detail, the resource allocation device includes a usage amount calculation unit that calculates a fixed usage amount, which is an actual usage amount of a resource, for each time slot obtained by dividing a resource fluctuation period by a virtual machine, a spike detection unit that detects a spike in the fixed usage amount, an allocation amount calculation unit that calculates an allocation amount of a resource to be allocated to the i-th time slot based on the past fixed usage amount in the i-th time slot and the detection result of the past spike in the time slot included in a predetermined range before and after the i-th time slot, and an allocation amount setting unit that sets the allocation amount in a virtual machine monitor that controls the virtual machine. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-091109 A Summary of the Invention [Problem to be solved by the invention]

[0004] In systems that must guarantee real-time performance, such as in-vehicle systems, dynamically adjusting resource allocation time makes it difficult to verify the feasibility of the system, so there is room for improvement.

[0005] The present invention has been made in consideration of the above-mentioned facts, and aims to provide an information processing device, a vehicle, an information processing method, and an information processing program that are applicable to a system in which resource allocation time is adjustable and real-time performance is required. [Means for solving the problem]

[0006] In order to achieve the above object, the information processing device according to claim 1 includes a generating unit that generates a plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines; Including the start-up phase, A detection unit that detects a plurality of predetermined phases, and a setting unit that sets the allocation times of resources of the plurality of virtual machines so as to conform to a schedule that is predetermined for each of the phases based on a detection result of the detection unit. When the start-up phase is detected by the detection unit, the setting unit sets the allocated time only to the management virtual machine, or sets the allocated time of the management virtual machine to a time longer than the allocated time of the other virtual machines. .

[0007] According to the invention recited in claim 1, the generation unit generates a plurality of virtual machines including a management virtual machine that manages a plurality of virtual machines.

[0008] The detection unit detects a plurality of predetermined phases, and the setting unit sets the resource allocation times of the plurality of virtual machines so as to achieve a predetermined schedule for each phase based on the detection results of the detection unit. This makes it possible to change the resource allocation time for each phase. Furthermore, since the CPU allocation time is fixed in each phase and the scheduling is static, the real-time performance required for an in-vehicle system can be guaranteed. Therefore, it is possible to provide an information processing device that is applicable to a system that has adjustable resource allocation times and requires real-time performance.

[0009] Also The plurality of phases includes a startup phase, and the setting unit detects the start-up phase by the detection unit. When the startup phase is detected, the allocated time may be set only for the management virtual machine, or the allocated time for the management virtual machine may be set to a time longer than those for the other virtual machines, thereby enabling the management virtual machine having functions such as initialization necessary for each virtual machine to operate to be started up earlier.

[0010] Furthermore, the plurality of phases may include a normal phase, and when the detection unit detects the normal phase, the setting unit may set the allocated time of the plurality of virtual machines to a predetermined normal time, thereby making it possible to allocate to each virtual machine an allocated time required for normal operation.

[0011] In addition, the plurality of phases may include a sleep phase, and when the detection unit detects the sleep phase, the setting unit may set the allocated time of the management virtual machine to be longer than those of the other virtual machines, thereby shortening the time required to transition to the sleep phase and the time required to return from the sleep phase.

[0012] A generation unit that generates the plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines; Including the start-up phase, A detection unit that detects a plurality of predetermined phases, and a setting unit that sets the allocation times of resources of the plurality of virtual machines so as to conform to a schedule that is predetermined for each of the phases based on a detection result of the detection unit. and when the detection unit detects the startup phase, the setting unit sets the allocated time only to the management virtual machine, or sets the allocated time of the management virtual machine to a time longer than the allocated time of the other virtual machines. The information processing device may be mounted on a vehicle.

[0013] The computer also generates a plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines; Including the start-up phase, Detecting a plurality of predetermined phases, and setting resource allocation times for the plurality of virtual machines so as to achieve a predetermined schedule for each of the phases based on the detection results of the phases. and when the boot phase is detected, the allocated time is set only for the management virtual machine, or the allocated time for the management virtual machine is set to a longer time than the allocated time for the other virtual machines. The information processing method may be an information processing method for performing a process to

[0014] Furthermore, a method for generating a plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines in a computer, Including the start-up phase, Detecting a plurality of predetermined phases, and setting resource allocation times for the plurality of virtual machines so as to achieve a predetermined schedule for each of the phases based on the detection results of the phases. and when the boot phase is detected, the allocated time is set only for the management virtual machine, or the allocated time for the management virtual machine is set to a longer time than the allocated time for the other virtual machines. An information processing program for executing the process Effect of the Invention

[0015] As described above, according to the present invention, it is possible to provide an information processing device, a vehicle, an information processing method, and an information processing program that are applicable to a system in which resource allocation time is adjustable and real-time performance is required. [Brief description of the drawings]

[0016] [Figure 1] 1 is a diagram showing a vehicle equipped with a central ECU according to an embodiment of the present invention. [Diagram 2] 2 is a block diagram showing a schematic configuration of a central ECU according to the present embodiment. FIG. [Diagram 3] FIG. 2 is a functional block diagram showing the functions of a Hypervisor. [Figure 4] FIG. 13 is a diagram illustrating an example of scheduling of a CPU allocation time for each phase. [Diagram 5] 6 is a flowchart showing an example of a process flow when setting a CPU allocation time that is performed by the central ECU according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] An example of an embodiment of the present invention will be described in detail below with reference to the drawings. In this embodiment, a central ECU mounted on a vehicle will be described as an example of an information processing device. In this embodiment, Fig. 1 is a diagram showing a vehicle mounted with a central ECU (Electronic Control Unit) according to this embodiment, and Fig. 2 is a block diagram showing a schematic configuration of the central ECU according to this embodiment.

[0018] The central ECU 12 according to this embodiment is mounted on the vehicle 10 and controls various ECUs provided in the vehicle 10 in an integrated manner.

[0019] The central ECU 12 is provided with a CPU (Central Processing Unit), and in this embodiment, as an example, as shown in FIG. 2, there are a plurality of CPU COREs 14 (four CPU COREs 1 to 4 in the example of FIG. 2).

[0020] In this embodiment, a physical CPU CORE 14 is virtualized by a Hypervisor 16, which is software for virtualizing a computer, and a VM (Virtual Machine) 18 is generated as a virtual machine. In this embodiment, a plurality of VMs 18 are generated by the Hypervisor 16. FIG. 2 shows an example in which three VMs 18, VM0 to VM2, are generated as the plurality of VMs 18.

[0021] An OS (Operating System) 20 is placed on each VM 18, and an application (App) 22 runs on the OS 20. In Fig. 2, App1 and App2 run on OS1, App3 and App4 run on OS2, and App5 and App6 run on OS3.

[0022] In a general hypervisor, the function of managing each VM is included in the hypervisor itself. However, in the hypervisor 16 for an in-vehicle system as in this embodiment, in order to guarantee real-time performance, the function of the hypervisor 16 is reduced as much as possible, and the function of managing each VM 18 is arranged as one VM 18. In this embodiment, the VM0 functions as a management virtual machine that manages each VM 18, and hereinafter, the VM0 may be referred to as the overall management VM 18. In addition, when the overall management VM 18 is provided, the degree of coupling between the overall management VM 18 and each VM 18 increases, and they are in a dependent relationship.

[0023] In addition, since a plurality of VMs 18 are arranged in the hypervisor 16, by allocating CPU time to each VM 18, it is possible to make each VM 18 appear to be operating in parallel.

[0024] In order to operate each VM 18 efficiently, it is desirable to dynamically change the scheduling of the CPU time allocation as a resource. However, in an in-vehicle system, if the scheduling is changed dynamically, it becomes difficult to guarantee real-time performance.

[0025] Therefore, in the central ECU 12 according to the present embodiment, the scheduling of the CPU allocation time is changed for each phase, so that the scheduling is semi-dynamic. This makes it possible to apply appropriate scheduling for each phase. In addition, since the scheduling is static for each phase, real-time performance can be guaranteed.

[0026] Here, the functional configuration of the hypervisor 16 for changing the scheduling of the CPU allocation time for each phase will be described.

[0027] As shown in FIG. 3, the hypervisor 16 has the functions of a generation unit 24, a detection unit 26, and a setting unit .

[0028] The generation unit 24 generates and executes a plurality of VMs 18 that are virtualized from the physical CPU CORE 14. In this embodiment, as described above, three VMs 18, VM0 to VM2, are generated.

[0029] The detector 26 detects a plurality of phases for changing the CPU allocation time. In this embodiment, the detector 26 detects three phases: a startup phase, a normal phase, and a sleep phase.

[0030] The setting unit 28 sets the scheduling of the CPU allocation time by changing the CPU allocation time so that the CPU allocation time conforms to a predetermined schedule for each phase based on the detection result of the detection unit 26. In this embodiment, the scheduling of the CPU allocation time is changed for each of the startup phase, normal phase, and sleep phase.

[0031] Fig. 4 is a diagram showing an example of the scheduling of the CPU allocation time for each phase. Fig. 4 shows an example of the scheduling of the CPU allocation time for the startup phase, normal phase, and sleep phase. In each phase, Major Time Frame=1000 μs.

[0032] In the start-up phase, the overall management VM 18 (VM0 in FIG. 2), which has the initialization and functions necessary for each VM 18 to operate, is started up early, so that VM0 occupies the CPU time. That is, in the start-up phase in FIG. 4, VM0 is set to 1000 μs, VM1 and VM2 to 0 μs, and after the overall management VM 18 is initialized, the process moves to the normal phase. Note that, although FIG. 4 shows an example in which VM0 occupies the CPU time, this is not limiting, and the CPU allocation time of VM0 may be set to be longer than that of the other VMs 1 and 2.

[0033] In the normal phase, a CPU allocation time required for normal operation is set as a predetermined normal time to each VM 18. In the example of Fig. 4, 200 µs is allocated to VM0, and 400 µs is allocated to VM1 and VM2.

[0034] In the sleep phase, data saving requests are concentrated on VM0, which is the overall management VM 18 that collectively manages the non-volatile memory, so the CPU time allocated to the overall management VM 18 is increased and the sleep time is shortened. In the example of Figure 4, 700 μs is allocated to VM0, and 150 μs is allocated to VM1 and VM2 respectively.

[0035] Next, a specific process performed by the central ECU 12 according to the present embodiment configured as described above will be described. Fig. 5 is a flowchart showing an example of a process flow when setting a CPU allocation time performed by the central ECU 12 according to the present embodiment. The process in Fig. 5 starts, for example, when a power source of the vehicle, such as an ignition switch (not shown), is turned on.

[0036] In step 100, the hypervisor 16 sets the CPU allocation time for the startup phase, and proceeds to step 102. That is, when the vehicle power is turned on, the detection unit 26 detects the startup phase, and the setting unit 28 sets the CPU allocation time for the startup phase. Specifically, as shown in Fig. 4, the VM0, which is the overall management VM 18, is set to 1000 μs, and the VM1 and VM2 are set to 0 μs, so that the VM0, which is the overall management VM 18, occupies the CPU time.

[0037] In step 102, the hypervisor 16 judges whether or not the transition to the normal phase has occurred. For example, the judgment is made by judging whether or not the detection unit 26 has received a notification of the end of the boot phase from the overall management VM 18. The process waits until the judgment is affirmative, and then proceeds to step 104.

[0038] In step 104, the hypervisor 16 changes the CPU allocation time to the normal phase, and the process proceeds to step 106. That is, the setting unit 28 changes the CPU allocation time setting to the normal phase. Specifically, as shown in FIG 4, the CPU allocation time is changed to 200 μs for VM0, and to 400 μs for VM1 and VM2.

[0039] In step 106, the hypervisor 16 judges whether or not the transition to the sleep phase has occurred. For example, the judgment is made by judging whether or not the detection unit 26 detects the establishment of a predetermined condition for transitioning to the sleep phase. If the judgment is affirmative, the process proceeds to step 108, and if the judgment is negative, the process proceeds to step 112.

[0040] In step 108, the hypervisor 16 changes the CPU allocation time setting to the sleep phase, and the process proceeds to step 110. That is, the setting unit 28 changes the CPU allocation time setting to the sleep phase. Specifically, as shown in Fig. 4, the CPU allocation time for the overall management VM 18 is increased by setting 700 μs for VM0, which is the overall management VM 18, and 150 μs for VM1 and VM2, respectively, thereby shortening the time for transition to and return from sleep.

[0041] In step 110, the hypervisor 16 judges whether or not to return from sleep. For example, the judgment judges whether or not the detection unit 26 detects the establishment of a predetermined condition for returning from sleep. If the judgment is affirmative, the process returns to step 100 and repeats the above-mentioned processing, and if the judgment is negative, the process proceeds to step 112. Note that in this embodiment, an example is shown in which the process returns to step 100 and the start-up phase is entered after returning from the sleep phase, but this is not limiting, and the process may proceed to the normal phase after returning from the sleep phase. In this case, if the judgment in step 110 is affirmative, the process proceeds to step 104.

[0042] In step 112, the hypervisor 16 judges whether or not to end the process. For example, the judgment is made by judging whether or not the detection unit 26 has detected that the vehicle power source, such as an ignition switch (not shown), has been turned off. If the judgment is negative, the process proceeds to step 114, and if the judgment is positive, the series of processes is ended.

[0043] In step 114, the hypervisor 16 judges whether or not it is in the normal phase. If it is in the normal phase, the judgment is affirmative, and the process returns to step 106 to repeat the above-mentioned processing. On the other hand, if it is in the sleep phase, the judgment is negative, and the process returns to step 110 to repeat the above-mentioned processing.

[0044] By performing the processing in this manner, it is possible to change the scheduling of the CPU time allocated to each VM 18 for each phase, and to adjust the resource allocation time.

[0045] In addition, since the CPU time allocated to each phase is fixed and the scheduling is static, the real-time performance required by in-vehicle systems can be guaranteed.

[0046] In the above embodiment, an example having four CPU cores 14 will be described, but the present invention is not limited to this. For example, a single CPU core 14 may be provided, or a number of CPU cores other than four may be provided.

[0047] In the above embodiment, the hypervisor 16 generates three VMs 18. However, the present invention is not limited to this. For example, the hypervisor 16 may generate two VMs 18, or four or more VMs 18.

[0048] In the above embodiment, the three phases, the start-up phase, the normal phase, and the sleep phase, are described as examples of phases, but the phases are not limited to these three phases. For example, there may be a plurality of different phases other than these three phases, or there may be two of the above three phases, or there may be a plurality of phases including the above three phases and another phase.

[0049] Furthermore, the processes performed by the hypervisor 16 in the above embodiment may be stored as a program in various storage media and distributed.

[0050] Furthermore, the present invention is not limited to the above, and it goes without saying that various modifications can be made without departing from the spirit and scope of the present invention. [Explanation of symbols]

[0051] 10 Vehicles 12 Central ECU (information processing unit) 14 CPU CORE 16 Hypervisor 18 VM 24 Generation part 26 Detection unit 28 Setting section

Claims

1. a generation unit that generates a plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines; A detection unit that detects a plurality of predetermined phases including a startup phase; a setting unit that sets resource allocation times of the plurality of virtual machines based on a detection result of the detection unit so as to conform to a predetermined schedule for each of the phases; Including, The information processing apparatus, wherein when the startup phase is detected by a detection unit, the setting unit sets the allocated time only to the management virtual machine, or sets the allocated time of the management virtual machine to a longer time than the allocated time of the other virtual machines.

2. the plurality of phases includes a normal phase; The information processing apparatus according to claim 1 , wherein the setting unit sets the allocated times of the virtual machines to a predetermined normal time when the normal phase is detected by a detection unit.

3. the plurality of phases includes a sleep phase; The information processing apparatus according to claim 1 , wherein the setting unit sets the allocated time of the management virtual machine to a longer time than the allocated time of the other virtual machines when the sleep phase is detected by the detection unit.

4. A generation unit that generates a plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines; A detection unit that detects a plurality of predetermined phases including a startup phase; a setting unit that sets resource allocation times of the plurality of virtual machines based on a detection result of the detection unit so as to conform to a predetermined schedule for each of the phases; wherein, when the startup phase is detected by a detection unit, the setting unit sets the allocated time only to the management virtual machine, or sets the allocated time of the management virtual machine to a longer time than the allocated time of the other virtual machines.

5. A computer comprising: generating a plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines; Detecting a plurality of predetermined phases including a start-up phase; An information processing method comprising: setting resource allocation times for the multiple virtual machines based on the detection result of the phase so as to conform to a predetermined schedule for each phase; and, when the startup phase is detected, setting the allocation time only for the management virtual machine, or setting the allocation time for the management virtual machine to a time longer than that for the other virtual machines.

6. A computer comprising: generating a plurality of virtual machines including a management virtual machine that manages the plurality of virtual machines; Detecting a plurality of predetermined phases including a start-up phase; An information processing program for executing a process of setting resource allocation time for the multiple virtual machines based on a result of the detection of the phase so as to conform to a predetermined schedule for each phase, and, when the startup phase is detected, setting the allocation time only for the management virtual machine, or setting the allocation time for the management virtual machine to a longer time than for the other virtual machines.

Citation Information

Patent Citations

  • Process scheduler using adaptive partitioning of process thread

    JP2006260565A

  • Address allocation method, computer, and program

    JP2010147929A

  • Resource allocation apparatus, method, and program

    JP2016091109A