Dynamic priority semi-perspective inexact mixed critical tasks energy consumption optimization scheduling method

Abstract

The present invention discloses a method for optimizing energy consumption scheduling of dynamic priority semi-perspective non-precise mixed critical tasks, which includes: establishing a dynamic priority semi-perspective mixed critical task graceful degradation scheduling model; determining scheduling rules and system correctness rules according to preset rules; obtaining the low-criticality level task τ according to a preset formula; i Processor requirements, high-criticality tasks τ j Processor requirements; according to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The present invention schedules mixed-critical tasks through a dynamic priority method and calculates the processor speed based on energy consumption using system correctness rules, which not only effectively improves system utilization but also reduces system energy consumption.

Description

Technical Field

[0001] The present invention relates to the field of real-time scheduling technology, in particular to a dynamic priority semi-perspective imprecise hybrid critical task energy consumption optimization scheduling method. Background Art

[0002] Mixed-criticality systems integrate applications of varying criticality onto a shared platform to meet system size, weight, and energy requirements. Graceful degradation means the system switches to a higher-criticality mode while still providing degraded services for lower-criticality tasks.

[0003] Mixed-criticality systems need to balance different mission requirements in terms of real-time performance and criticality, and typically have their own defined safety levels. For example, the avionics safety certification standard DO-178C defines five criticality levels (AE), with A being the highest and E being the lowest. Standards in fields such as automotive electronics also clearly define the grading of different system functions.

[0004] In the aviation field, drones are usually battery-powered and need to operate in low-end environments, so their flight time is particularly important. Low energy consumption can not only improve battery life, but also increase processor life and system reliability, thereby enhancing product competitiveness.

[0005] Existing technologies for mixed-criticality energy optimization primarily focus on the classic non-perspective mixed-criticality model. In this model, low-criticality tasks are directly discarded in high-criticality mode, ignoring the conditions under which these tasks can continue to execute. This wastes system resources and is impractical, so the corresponding algorithms cannot be directly applied to drones. Furthermore, existing technologies primarily determine the timing of system mode switching by determining whether the execution time (WorstCase Execute Time, WCET) of the high-criticality layer exceeds its worst-case execution time in the low-criticality mode. This results in low system utilization.

[0006] Therefore, there are problems in the prior art such as waste of system resources and low system utilization. Summary of the Invention

[0007] The main purpose of the present invention is to provide a dynamic priority semi-perspective non-precise hybrid critical task energy consumption optimization scheduling method, which aims to solve the technical problems of system resource waste and low system utilization in the existing technology.

[0008] To achieve the above objectives, the present invention provides a dynamic priority semi-perspective non-precise mixed critical task energy consumption optimization scheduling method, which includes the following steps: establishing a dynamic priority semi-perspective mixed critical task graceful degradation scheduling model; determining the scheduling rules and system correctness rules according to preset rules; obtaining the low-criticality level task τ according to a preset formula i Processor requirements, high-criticality tasks τ j Processor requirements; according to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The processor demand is calculated, the processor speed is calculated, and scheduling is performed at that processor speed.

[0009] Optionally, a dynamic priority semi-perspective mixed-criticality task graceful degradation scheduling model is established, specifically including: in a single processor, there is a set of n mixed-criticality occasional tasks with deadline constraints Γ={τ1,τ2,…,τ n}, and use the preemptive dynamic priority strategy to schedule the task set; mixed critical occasional tasks τ i By the quaternion {T i ,D i ,ξ i ,C i}, and 1≤i≤n, i is an integer; where T i For mixed-criticality occasional tasks τ i The cycle of D i For mixed-criticality occasional tasks τ i The relative deadline of D i ≤T i ξ i For mixed-criticality occasional tasks τ i The key level of i ∈{LO,HI}, when mixed critical occasional tasks τ i When the key level is LO, that is, ξ i =LO, τ i For low-criticality tasks, when mixed-criticality occasional tasks τ i When the key level is HI, that is, ξ i =HI, τ i For high-criticality tasks; C i represents a mixed-criticality occasional task τ i The worst execution time in different modes, and C i ∈{C i (LO), C i (HI)}, when C i =C i (LO) indicates a mixed-criticality occasional task τ iThe worst execution time in low mode, when C i =C i (HI) indicates a mixed-criticality occasional task τ i Worst-case execution time in high mode.

[0010] Optionally, the scheduling rule and the system correctness rule are determined according to the preset rule, specifically including: determining the first preset rule as the scheduling rule, the first preset rule is specifically: the system is in low mode at the beginning; when the high-criticality task τ j At time t s Arrives and triggers mode conversion, at which time the system switches from low mode to high mode, t s is the moment of system mode conversion, and the release moment is at time t s Previous high-criticality level tasks τ j The execution time does not exceed τ j Worst case execution time in low mode C j (LO); release time is at time t s and time t s Later high-criticality tasks τ j The execution time of C j (LO) Upgrade to C j (HI), where C j (HI) is τ j The worst execution time in high mode; at time t s Low-criticality tasks that were not completed before τ i Or at time t s Released low-criticality tasks τ i The execution time is C i (LO); at time t s Low-criticality tasks τ that will be released later i The execution time is C i (HI); when the processor is idle or all high-criticality tasks are completed, the system switches from high mode to low mode; the second preset rule is determined as the system correctness rule, and the second preset rule is specifically: regardless of the mode the system is in, all tasks are completed within their corresponding deadlines; that is, within any time interval t, the processor demand of task set Γ in low mode and high mode is less than or equal to the time interval t, and the value of t is less than or equal to the super period of task set Γ.

[0011] Optionally, the super period of the task set Γ is equal to the least common multiple of all task periods.

[0012] Optionally, the system is in low mode if any task τ i The execution time does not exceed τ iWorst case execution time in low mode C i (LO); the system is in high mode when any task τ i The execution time does not exceed τ i The worst execution time in high mode C i (HI); If the mixed critical occasional task τ i For low-criticality tasks, C i (HI) <C i (LO); If the mixed-criticality occasional task τ i For high-criticality tasks, C i (HI)≥C i (LO).

[0013] Optionally, a dynamic priority strategy means that the priority of a task changes during execution. Specifically, the dynamic priority strategy uses an earliest deadline first strategy to schedule sporadic task sets.

[0014] Optionally, the earliest deadline first scheduling strategy is specifically to use the task τ i The absolute deadline of the task determines the priority; i The smaller the absolute deadline of task τ, the higher its priority; if task τ i The absolute deadline of task τ is the same as i The earlier the arrival time of task τ, the higher its priority; if task τ i The absolute deadline and arrival time of task τ are the same. i The smaller the subscript i is, the higher the priority is, and the high-priority task will be executed first; the task τ i The absolute deadline of is equal to the sum of its arrival time and the relative deadline.

[0015] Optionally, the preset formula includes a first preset formula, a second preset formula, a third preset formula and a fourth preset formula; low-key level task τ i The processor requirements include low-criticality tasks τ i Processor requirements in low mode dbf LO (τ i ,t) and low-criticality tasksτ i Processor requirements in high mode dbf HI (τ i ,t s ,t); high-criticality tasksτ j The processor requirements include high criticality level tasks τ j Processor requirements in low mode dbf LO (τ j ,t) and high-criticality tasksτ j Processor requirements in high mode dbfHI (τ j ,t s ,t); According to the preset formula, the low-key level task τ is obtained i Processor requirements, high-criticality tasks τ j The processor requirements of the low-criticality level task τ are specifically obtained according to the first preset formula. i Processor requirements in low mode dbf LO (τ i ,t), according to the second preset formula, we get the low-key level task τ i Processor requirements in high mode dbf HI (τ i ,t s ,t); According to the third preset formula, the high-key level task τ is obtained j Processor requirements in low mode dbf LO (τ j ,t); According to the fourth preset formula, the high-key level task τ is obtained j Processor requirements in high mode dbf HI (τ j ,t s ,t).

[0016] Optionally, the first preset formula is specifically: dbf LO (τ i ,t)=ψ i (t)C i (LO) / S; where ψ i (t) is the task τ within the time interval t i The number of jobs released, and S is the processor speed;

[0017] The second preset formula is: dbf HI (τ i ,t s ,t)=ψ i (t)C i (HI) / S+b i (C i (LO)-C i (HI)) / S; where b i To release the moment at t s Later, within the time interval t, the task τ i The number of jobs released, and

[0018] The third preset formula is: dbf LO (τ j ,t)=ψ j (t)C j(LO) / S; where ψ j (t) is the task τ within the time interval t j The number of jobs released, and S is the processor speed;

[0019] The fourth preset formula is: dbf HI (τ j ,t s ,t)=ψ j (t)C j (LO) / S+a j (C j (HI)-C j (LO)) / S; where a j To release the moment at t s Later, within the time interval t, the task τ j The number of jobs released, and a j =ψ j (tt s ).

[0020] Optionally, according to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The processor requirements are calculated to calculate the processor speed, including: according to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The processor requirements of the system in low mode should meet the following formula:

[0021]

[0022] That is, according to the low-criticality level task τ i Processor requirements in low mode dbf LO (τ i ,t), high-criticality tasksτ j Processor requirements in low mode dbf LO (τ j ,t) The relationship between the sum of the two and the time interval t is used to calculate the processor speed;

[0023] According to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The processor requirements of the system in high mode meet the following equation:

[0024]

[0025] That is, according to the low-criticality level task τ iProcessor requirements in high mode dbf HI (τ i ,t s ,t), high-criticality tasksτ j Processor requirements in high mode dbf HI (τ j ,t s ,t) and the relationship between the sum of the two and the time interval t, the processor speed is calculated; where, Γ LO is the set of low-criticality tasks, Γ HI A collection of high-criticality tasks.

[0026] The beneficial effects of the present invention are:

[0027] (1) Scheduling mixed-criticality tasks through a dynamic priority method and calculating the processor speed of energy consumption using system correctness rules can not only effectively improve system utilization but also reduce system energy consumption;

[0028] (2) The method provided by the present invention can provide degraded services for non-critical tasks when the system is in high mode, thereby improving the utilization of system resources;

[0029] (3) Compared with other algorithms, the method provided by the present invention can save 50.07% of normalized energy consumption, thereby reducing the energy consumption of the hybrid critical system and ensuring the real-time performance and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0031] Figure 1 This is a simplified flow chart of the dynamic priority semi-perspective non-precise mixed critical task energy consumption optimization scheduling method of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0033] like Figure 1As shown in the figure, a dynamic priority semi-perspective non-precise mixed critical task energy consumption optimization scheduling method of the present invention includes the following steps: establishing a dynamic priority semi-perspective mixed critical task graceful degradation scheduling model; determining the scheduling rules and system correctness rules according to preset rules; obtaining the low-criticality level task τ according to a preset formula i Processor requirements, high-criticality tasks τ j Processor requirements; according to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The processor demand is calculated, the processor speed is calculated, and scheduling is performed at that processor speed.

[0034] It's important to note that "semi-perspective" is a technical term that primarily refers to a scheduling mechanism. The semi-perspective mixed-criticality model differs from the traditional non-perspective mixed-criticality model primarily in its system mode transition mechanism. In the semi-perspective mixed-criticality model, the system can determine whether to switch from low to high mode upon the arrival of a high-criticality task. In the traditional non-perspective mixed-criticality model, the system waits until a high-criticality task begins executing and its execution time exceeds its worst-case execution time in low mode before switching from low to high mode.

[0035] Graceful degradation is mainly aimed at low-criticality tasks in high mode. It means that when the system switches to high mode, it can still provide degraded services for low-criticality tasks, but the service time is less than the worst-case execution time in low mode.

[0036] The present invention schedules mixed critical tasks through a dynamic priority method and calculates the processor speed of energy consumption using a system correctness rule, which not only effectively improves the utilization rate of the system but also reduces the energy consumption of the system.

[0037] In this embodiment, a dynamic priority semi-perspective mixed-criticality task graceful degradation scheduling model is established, specifically including: in a single processor, there are n deadline-constrained mixed-criticality occasional task sets Γ={τ1,τ2,…,τ n}, and use the preemptive dynamic priority strategy to schedule the task set; mixed critical occasional tasks τ i By the quaternion {T i ,D i ,ξ i ,C i}, and 1≤i≤n, i is an integer; where T i For mixed-criticality occasional tasks τ i The cycle of D i For mixed-criticality occasional tasks τ i The relative deadline of D i ≤T iξ i For mixed-criticality occasional tasks τ i The key level of i ∈{LO,HI}, when mixed critical occasional tasks τ i When the key level is LO, that is, ξ i =LO, τ i For low-criticality tasks, when mixed-criticality occasional tasks τ i When the key level is HI, that is, ξ i =HI, τ i For high-criticality tasks; C i represents a mixed-criticality occasional task τ i The worst execution time in different modes, and C i ∈{C i (LO), C i (HI)}, when C i =C i (LO) indicates a mixed-criticality occasional task τ i The worst execution time in low mode, when C i =C i (HI) indicates a mixed-criticality occasional task τ i Worst-case execution time in high mode.

[0038] In this embodiment, the scheduling rule and the system correctness rule are determined according to the preset rules, specifically including: determining the first preset rule as the scheduling rule, the first preset rule is specifically: the system is in low mode at the beginning; when the high-criticality task τ j At time t s Arrives and triggers mode conversion, at which time the system switches from low mode to high mode, t s is the moment of system mode conversion, and the release moment is at time t s Previous high-criticality level tasks τ j The execution time does not exceed τ j Worst case execution time in low mode C j (LO); release time is at time t s and time t s Later high-criticality tasks τ j The execution time of C j (LO) Upgrade to C j (HI), where C j (HI) is τ j The worst execution time in high mode; at time t s Low-criticality tasks that were not completed before τ i Or at time t s Released low-criticality tasks τ iThe execution time is C i (LO); at time t s Low-criticality tasks τ that will be released later i The execution time is C i (HI); when the processor is idle or all high-criticality tasks are completed, the system switches from high mode to low mode; the second preset rule is determined as the system correctness rule, and the second preset rule is specifically: regardless of the mode the system is in, all tasks are completed within their corresponding deadlines; that is, within any time interval t, the processor demand of task set Γ in low mode and high mode is less than or equal to the time interval t, and the value of t is less than or equal to the super period of task set Γ.

[0039] Preferably, the super period of the task set Γ is equal to the least common multiple of all task periods.

[0040] In this embodiment, a task can release multiple jobs. The release time refers to the release time of the task. For example, the release time of the first job is 0, and the release time of the second job is equal to 0 plus the period of the second job; and so on. That is, the release time of the subsequent job is equal to the release time of the previous job plus the period.

[0041] In this embodiment, the system is in low mode when any task τ i The execution time does not exceed τ i Worst case execution time in low mode C i (LO); the system is in high mode when any task τ i The execution time does not exceed τ i The worst execution time in high mode C i (HI); If the mixed critical occasional task τ i For low-criticality tasks, C i (HI) <C i (LO); If the mixed critical occasional task τ i For high-criticality tasks, C i (HI)≥C i (LO).

[0042] In this embodiment, the dynamic priority policy means that the priority of a task changes during execution. The dynamic priority policy preferably uses the earliest deadline first policy to schedule sporadic task sets.

[0043] In this embodiment, the earliest deadline priority scheduling strategy is to use the task τ i The absolute deadline of the task determines the priority; i The smaller the absolute deadline of task τ, the higher its priority; if task τ iThe absolute deadline of task τ is the same as i The earlier the arrival time of task τ, the higher its priority; if task τ i The absolute deadline and arrival time of task τ are the same. i The smaller the subscript i is, the higher the priority is, and the high-priority task will be executed first; the task τ i The absolute deadline of is equal to the sum of its arrival time and the relative deadline.

[0044] In this embodiment, the preset formula includes a first preset formula, a second preset formula, a third preset formula and a fourth preset formula; the low-key level task τ i The processor requirements include low-criticality tasks τ i Processor requirements in low mode dbf LO (τ i ,t) and low-criticality tasksτ i Processor requirements in high mode dbf HI (τ i ,t s ,t); high-criticality tasksτ j The processor requirements include high criticality level tasks τ j Processor requirements in low mode dbf LO (τ j ,t) and high-criticality tasksτ j Processor requirements in high mode dbf HI (τ j ,t s ,t); According to the preset formula, the low-key level task τ is obtained i Processor requirements, high-criticality tasks τ j The processor requirements of the low-criticality level task τ are specifically obtained according to the first preset formula. i Processor requirements in low mode dbf LO (τ i ,t), according to the second preset formula, we get the low-key level task τ i Processor requirements in high mode dbf HI (τ i ,t s ,t); According to the third preset formula, the high-key level task τ is obtained j Processor requirements in low mode dbf LO (τ j ,t); According to the fourth preset formula, the high-key level task τ is obtained j Processor requirements in high mode dbf HI (τ j ,t s ,t).

[0045] Preferably, the first preset formula is specifically:

[0046] dbf LO (τ i ,t)=ψ i (t)C i (LO) / S;

[0047] Among them, ψ i (t) is the task τ within the time interval t i The number of jobs released, and S is the processor speed.

[0048] Preferably, the second preset formula is specifically:

[0049] dbf HI (τ i ,t s ,t)=ψ i (t)C i (HI) / S+b i (C i (LO)-C i (HI)) / S;

[0050] Among them, b i To release the moment at t s Later, within the time interval t, the task τ i The number of jobs released, and

[0051] Preferably, the third preset formula is specifically:

[0052] dbf LO (τ j ,t)=ψ j (t)C j (LO) / S;

[0053] Among them, ψ j (t) is the task τ within the time interval t j The number of jobs released, and

[0054] Preferably, the fourth preset formula is specifically:

[0055] dbf HI (τ j ,t s ,t)=ψ j (t)C j (LO) / S+a j (C j (HI)-C j (LO)) / S;

[0056] Among them, a j To release the moment at t s Later, within the time interval t, the task τ j The number of jobs released, and a j =ψ j (tt s ).

[0057] In this embodiment, according to the system correctness rule, the low-criticality level task τ i Processor requirements, high-criticality tasks τ j The processor requirements are calculated to calculate the processor speed, including: according to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The processor requirements of the system in low mode should meet the following formula:

[0058]

[0059] That is, according to the low-criticality level task τ i Processor requirements in low mode dbf LO (τ i ,t), high-criticality tasksτ j Processor requirements in low mode dbf LO (τ j ,t) The relationship between the sum of the two and the time interval t is used to calculate the processor speed;

[0060] According to the system correctness rules, low-criticality tasks τ i Processor requirements, high-criticality tasks τ j The processor requirements of the system in high mode meet the following equation:

[0061]

[0062] That is, according to the low-criticality level task τ i Processor requirements in high mode dbf HI (τ i ,t s ,t), high-criticality tasksτ j Processor requirements in high mode dbf HI (τ j ,t s ,t) and the relationship between the sum of the two and the time interval t, the processor speed S is calculated; where, Γ LO is the set of low-criticality tasks, Γ HI A collection of high-criticality tasks.

[0063] The present invention schedules mixed critical tasks through a dynamic priority method and calculates the processor speed of energy consumption using a system correctness rule, which not only effectively improves the utilization rate of the system but also reduces the energy consumption of the system.

[0064] The method provided by the present invention can provide degraded services for non-critical level tasks when the system is in high mode, thereby improving the utilization rate of system resources.

[0065] In this embodiment, the mixed sporadic task set Γ = {τ1, τ2} contains two sporadic tasks; the period T1 of the sporadic task τ1 is equal to 5, its relative deadline D1 is 5, and its critical level ξ1 is LO, that is, it is a low-criticality task, its low-mode worst-case execution time C1(LO) is 2; its high-mode worst-case execution time C1(HI) is 1; the period T2 of the sporadic task τ2 is equal to 20, its relative deadline D2 is 20, and its critical level ξ2 is HI, that is, it is a high-criticality task, its low-mode worst-case execution time C2(LO) is 6; its high-mode worst-case execution time C2(HI) is 8; assuming that the first job of τ1 is released at time 0 and the first job of τ2 is released at time 1; the super period of the task set is 20; since only high-criticality tasks can trigger mode switching, if a mode switching occurs, it must be at time 1, that is, t s =1; through calculation, we can know that the speed of processor energy consumption optimization is S=0.7.

[0066] The normalized energy consumption of the method of the present invention in low mode is 0.353, while the normalized energy consumption of other methods in low mode is 0.707. Therefore, compared with other algorithms, the method provided by the present invention saves 50.07% of normalized energy consumption; it reduces the energy consumption of the hybrid critical system and ensures the real-time and reliability of the system.

[0067] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0068] While the foregoing description shows and describes preferred embodiments of the present invention, it should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. Rather, the present invention can be used in various other combinations, modifications, and environments, and can be modified within the scope of the present invention by the teachings herein or by techniques or knowledge in the relevant art. Modifications and variations made by those skilled in the art without departing from the spirit and scope of the present invention are intended to be within the scope of the appended claims.

Claims

1. A dynamic priority semi-perspective non-precise mixed critical task energy consumption optimization scheduling method, characterized by: The following steps are involved: Establish a dynamic priority semi-perspective mixed critical task graceful degradation scheduling model; Determine the scheduling rules and system correctness rules respectively according to the preset rules; Get low-key level tasks according to the preset formula Processor requirements, high-criticality tasks Processor requirements; According to the system correctness rules, low-criticality tasks Processor requirements and high-criticality tasks The processor demand, calculates the processor speed, and schedules at that processor speed; A dynamic priority, semi-perspective, mixed-criticality graceful degradation scheduling model is established, specifically including: In a single processor, there are n deadline-constrained mixed-criticality sporadic tasks. , and use the preemptive dynamic priority strategy to schedule the task set; Mixed-criticality occasional tasks By quad Composed of, and 1≤i≤n, i is an integer; For mixed-criticality occasional tasks cycle; For mixed-criticality occasional tasks The relative deadline of ; For mixed-criticality occasional tasks The key level, and ∈ , when mixed critical occasional tasks The key levels are When =LO, For low-criticality tasks, when mixed-criticality occasional tasks The key levels are When =HI, For high-criticality tasks; Represents mixed-criticality occasional tasks The worst execution time in different modes, and ∈{ , },when = When , it indicates a mixed-criticality occasional task The worst execution time in low mode is = When , it indicates a mixed-criticality occasional task Worst execution time in high mode; Determine the scheduling rules and system correctness rules based on the preset rules, including: The first preset rule is determined as the scheduling rule. The first preset rule is specifically: the system is in low mode at the beginning; when the high-key level task At the moment Arrives and triggers mode conversion, at which time the system switches from low mode to high mode, The moment of system mode conversion, the release moment is at the moment Previous high-criticality tasks The execution time does not exceed Worst execution time in low mode ; Release the moment at the moment and time Future high-key tasks The execution time from Upgrade to ,in for Worst execution time in high mode; at time Low-criticality tasks that were not previously completed Or at the moment Released low-criticality tasks The execution time is ; at the moment Low-criticality tasks to be released later The execution time is When the processor is idle or all high-criticality tasks have completed execution, the system switches from high mode to low mode. The second preset rule is determined as the system correctness rule. The second preset rule is specifically: no matter what mode the system is in, all tasks are completed within their corresponding deadlines; that is, at any time interval In the task set Processor demand in both low and high modes is less than or equal to the time interval , The value is less than or equal to the task set The super cycle; The preset formulas include a first preset formula, a second preset formula, a third preset formula, and a fourth preset formula; Low-criticality tasks Processor requirements include low-criticality tasks Processor requirements in low mode and low-criticality tasks Processor requirements in high mode ; High-criticality tasks Processor requirements include high-criticality tasks Processor requirements in low mode and high-criticality tasks Processor requirements in high mode ; Get low-key level tasks according to the preset formula Processor requirements, high-criticality tasks Processor requirements, including: According to the first preset formula, the low-key level tasks are obtained Processor requirements in low mode , according to the second preset formula, we can get the low-key level tasks Processor requirements in high mode ;According to the third preset formula, high-key level tasks are obtained Processor requirements in low mode ;According to the fourth preset formula, high-key level tasks are obtained Processor requirements in high mode ; The first preset formula is: ;in, For the time interval Internal Tasks The number of jobs released, and , S is the processor speed; The second preset formula is: ;in, To release the moment Later, the time interval Internal Tasks The number of jobs released, and ; The third preset formula is: ;in, For the time interval Internal Tasks The number of jobs released, and ; The fourth preset formula is specifically: ;in, To release the moment Later, the time interval Internal Tasks The number of jobs released, and ; According to the system correctness rules, low-criticality tasks Processor requirements and high-criticality tasks Processor requirements, calculate processor speed, including: According to the system correctness rules, low-criticality tasks Processor requirements and high-criticality tasks The processor requirements of the system in low mode should meet the following formula: ； That is, according to the low-criticality level tasks Processor requirements in low mode , high-criticality tasks Processor requirements in low mode The relationship between the sum of the two and the time interval t is used to calculate the processor speed; According to the system correctness rules, low-criticality tasks Processor requirements and high-criticality tasks The processor requirements of the system in high mode meet the following equation: ； That is, according to the low-criticality level tasks Processor requirements in high mode , high-criticality tasks Processor requirements in high mode The relationship between the sum of the two and the time interval t is used to calculate the processor speed; in, is a set of low-criticality tasks, A collection of high-criticality tasks.

2. The dynamic priority semi-transparent imprecise mixed critical task energy consumption optimization scheduling method according to claim 1 is characterized by: Task Set The super period is equal to the least common multiple of all task periods.

3. The dynamic priority semi-transparent imprecise mixed critical task energy consumption optimization scheduling method according to claim 1 is characterized by: The system is in low mode when any task The execution time does not exceed Worst execution time in low mode ; The system is in high mode when any task The execution time does not exceed Worst execution time in high mode ; If mixed critical occasional tasks For low-criticality tasks, ; If mixed critical occasional tasks For high-criticality tasks, .

4. The method for energy consumption optimization scheduling of dynamic priority semi-transparent imprecise mixed critical tasks according to claim 1 is characterized by: The dynamic priority strategy means that the priority of a task changes during execution. Specifically, the dynamic priority strategy uses the earliest deadline first strategy to schedule sporadic task sets.

5. The dynamic priority semi-transparent imprecise mixed critical task energy consumption optimization scheduling method according to claim 4 is characterized by: The earliest deadline first scheduling strategy is to Prioritize by absolute deadline; Task The smaller the absolute deadline of , the higher its priority; If the task have the same absolute deadline, then the task The earlier the arrival time, the higher its priority; If the task The absolute deadline and arrival time of task The smaller the subscript i is, the higher the priority is, and high-priority tasks are executed first; Task The absolute deadline of is equal to the sum of its arrival time and the relative deadline.

Images