Distribution master station computing power resource adaptive scheduling method and device

Abstract

The invention provides a power distribution master station computing power resource adaptive scheduling method and device, and the method comprises the steps: monitoring the dynamic change of the to-be-decompressed data size of each business in real time according to a preset time section division strategy, and obtaining a computing power resource distribution scheme through the analysis of a preset constraint condition according to the change condition; calculating the decompression duration of each service in the current time section according to the computing power resource allocation scheme, accumulating the decompression durations to obtain the total decompression duration of each service, and calculating the delay redundancy rate and the timeout percentage; according to the Markoviz mean variance model, a network supporting force objective function is constructed through the time delay redundancy rate and the timeout percentage, and according to the network supporting force, the objective function is constructed with the maximization of the network supporting force as a target; and obtaining an optimized computing power resource allocation parameter and a time section division strategy through objective function analysis, dynamically adjusting the resources of the power distribution master station according to the computing power resource allocation parameter, and triggering time section reset so as to reallocate idle resources.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of power distribution system communication, and in particular to a power distribution master station computing resource adaptive scheduling method and device. BACKGROUND

[0002] With the rapid development of smart grids and energy internet, a large amount of real-time monitoring data is generated by the sensing devices (such as PMU, smart meters, fault indicators, etc.) deployed in the power distribution network, including high-frequency sampling information of voltage, current, power, and harmonics. These data are the basis for state estimation, fault diagnosis, and optimal scheduling, but due to the limitations of communication bandwidth and storage cost, the original data usually needs to be compressed before being transmitted to the edge server or cloud for processing. However, traditional compression algorithms (such as Huffman coding in lossless compression or wavelet transform in lossy compression) face two major challenges in the power distribution network scenario: first, the time-space correlation of high-frequency data is not fully exploited, making it difficult to balance compression rate and accuracy; second, the decompression process requires a large amount of computing resources, and the computing resources of the edge nodes in the power distribution network are limited, so the computing resources need to be dynamically allocated to balance real-time performance and energy efficiency.

[0003] In terms of computing resource allocation, traditional static allocation strategies cannot adapt to the dynamic fluctuations of power distribution network load (such as data surge during fault events). In recent years, dynamic resource scheduling methods based on reinforcement learning (RL) or federated learning (FL) have been proposed, which perceive the network topology, data priority, and node computing state to achieve collaborative optimization of decompression tasks and computing resources. For example, GPU acceleration can be preferentially allocated for critical fault data decompression, while historical archive data can be processed using CPU batch processing. In addition, joint optimization of computing and communication has also become a research hotspot, such as deploying lightweight decompression models (such as pruned neural networks) on the edge side, and only uploading feature vectors to the cloud for reconstruction to reduce transmission overhead.

[0004] Therefore, there is an urgent need in the industry for a scheduling method to solve some or all of the above technical problems. SUMMARY

[0005] The application aims to provide a power distribution master station computing resource adaptive scheduling method and device to dynamically adjust the computing resource allocation scheme by monitoring the changes in the amount of data to be decompressed in real time, improve resource utilization and reduce the risk of timeout, introduce a variable step mechanism to dynamically adjust the length and number of time sections according to the business decompression progress, reduce resource fragmentation; combine the time delay redundancy rate and the timeout percentage to build a network support force objective function to balance the real-time requirements and resource efficiency; design a distributed asynchronous collaboration strategy to trigger resource reallocation immediately when the business decompression is completed, and realize on-demand recovery and reallocation of computing resources.

[0006] To achieve the above purpose, the power distribution master station computing resource adaptive scheduling method provided by the application comprises: monitoring the dynamic change of the amount of each service to be decompressed in real time according to a preset time section division strategy, and obtaining a computing resource allocation scheme through a preset constraint condition according to the change of the amount of data to be decompressed; calculating the decompression time length of each service in the current time section according to the computing resource allocation scheme, accumulating the decompression time length of each time section to obtain the total decompression time length of each service, and calculating the time delay redundancy rate and the timeout percentage based on the total decompression time length; constructing a network support force objective function through the time delay redundancy rate and the timeout percentage according to the Markowitz mean-variance model, and constructing a target function according to the network support force with the goal of maximizing the network support force; obtaining optimized computing resource allocation parameters and time section division strategies through the target function analysis, dynamically adjusting the power distribution master station resources according to the optimized computing resource allocation parameters, and resetting the time section when the service decompression is completed to re-allocate idle resources.

[0007] In the above power distribution master station computing resource adaptive scheduling method, optionally, the constraint condition comprises a first constraint condition and a second constraint condition; the first constraint condition is that the total computing resource allocated to each service does not exceed the total computing resource of the power distribution master station; and the second constraint condition is that the computing resource allocated to each service is non-negative and does not exceed the maximum demand value.

[0008] In the above power distribution master station computing resource adaptive scheduling method, optionally, calculating the decompression time length of each service in the current time section according to the computing resource allocation scheme comprises: calculating the decompression time length of each service according to the number of processor cycles required for decompression unit bits, the amount of data to be decompressed, a preset minimum constant and a time section constant through the following formula:

[0009]

[0010] In the above formula, ρ k is the number of CPU cycles required for decompression unit bits of service k, D k,i is the amount of data to be decompressed by service k at the beginning of time section i, is a minimum constant, which prevents the problem from being solved when f k,i is 0, t i is the length of time section i, t k,i is the decompression time length of service k.

[0011] In the above power distribution master station computing resource adaptive scheduling method, optionally, constructing a network support force objective function through the time delay redundancy rate and the timeout percentage according to the Markowitz mean-variance model comprises:

[0012] The network support force objective function is constructed by the following formula:

[0013]

[0014] In the above formula, E (θ k ) is the average of the time delay redundancy rate θ k , E (γ k ) is the average of the service timeout percentage γ k , and E (C) is the network support force of all services in the network.

[0015] In the power distribution master station computing resource adaptive scheduling method, the target function is constructed according to the network support force with the maximum network support force as the target, which comprises:

[0016] The analysis model is constructed by the following formula:

[0017]

[0018] In the above formula, is the network support force of all services in the network; constraint C1 is the computing resource constraint, that is, the sum of the computing resources allocated by the power distribution master station to each service cannot exceed the total computing power; constraint C2 indicates that the computing resources allocated to each service are non-negative and cannot be greater than the total computing resources; constraint C3 is the time section quantity constraint, the number of time sections cannot exceed the total number of service types; constraint C4 is the time section length constraint, the sum of the lengths of each time section cannot exceed the total decompression time of the power distribution master station; k is the service set; δ is the number of time sections; t i is the length of time section i; f all is the total amount of computing resources of the power distribution master station; f k is the computing frequency allocated to each service.

[0019] In the power distribution master station computing resource adaptive scheduling method, the optimization computing resource allocation parameters are obtained by analyzing the target function, which comprises: constructing a Lagrange function according to the target function, obtaining the optimal solution of the target function based on the Lagrange function and the KKT condition; and obtaining the computing resource allocation parameters by iteratively solving the optimal solution of the target function.

[0020] The Lagrange function comprises:

[0021]

[0022] The KKT condition comprises:

[0023]

[0024] In the above formula, ε and η​k and χ k respectively corresponding to Lagrange multipliers of constraints C1 and C2; is a first derivative of , is an optimal solution of the computing resource allocated to each service, ε * , and are optimal solutions of ε, η k and χ k respectively corresponding to optimal solutions of ε, η and χ; is a network support force corresponding to the service k affected by the allocated computing resource f k .

[0025] In the power distribution master station computing resource adaptive scheduling method described above, the time section division strategy obtained by analyzing the objective function can be selected, which comprises: using a distributed asynchronous collaborative mechanism to analyze the objective function to obtain the time section division strategy.

[0026] In the power distribution master station computing resource adaptive scheduling method described above, the distributed asynchronous collaborative mechanism can comprise: when it is detected that any service is completed decompression, interrupting the current time section; releasing the computing resource occupied by the decompressed service, and dynamically adjusting the priority according to the time delay redundancy rate of the remaining service, and reallocating the computing resource; generating a new time section division scheme based on the updated to-be-decompressed data volume and the objective function; wherein the time section division step is dynamically adjusted according to the service type.

[0027] The application also provides a power distribution master station computing resource adaptive scheduling device, which comprises a monitoring module, a calculation module, a construction module and an analysis module; the monitoring module is used to monitor the dynamic change of the to-be-decompressed data volume of each service in real time according to a preset time section division strategy, and to obtain a computing resource allocation scheme by analyzing a preset constraint condition according to the change of the to-be-decompressed data volume; the calculation module is used to calculate the decompression time length of each service within the current time section according to the computing resource allocation scheme, to obtain the total decompression time length of each service by accumulating the decompression time length of each time section, and to calculate the time delay redundancy rate and the timeout percentage based on the total decompression time length; the construction module is used to construct a network support force objective function based on the time delay redundancy rate and the timeout percentage according to the Markowitz mean-variance model, and to construct an objective function with the maximum network support force as the target according to the network support force; and the analysis module is used to obtain optimized computing resource allocation parameters and a time section division strategy by analyzing the objective function, to dynamically adjust the power distribution master station resource according to the optimized computing resource allocation parameters, and to trigger time section reset when the service decompression is completed, so as to reallocate the idle resource.

[0028] The application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0029] The application further provides a computer readable storage medium, which stores a computer program for executing the above method.

[0030] The application further provides a computer program product, comprising computer programs / instructions, which, when executed by a processor, implement the steps of the above method.

[0031] The application has the beneficial technical effects that: by adaptively allocating computing power resources, the total computing power resource utilization rate of the power distribution master station is improved, especially in the data volume fluctuation scenario. The decompression timeout percentage of differentiated services is reduced, and the time delay redundancy rate of key services (such as fault alarms) is improved. The variable step size time section division strategy enables the system to quickly respond to changes in service demand, and the time section division granularity can be dynamically adjusted according to the load, reducing the computing overhead. The distributed asynchronous cooperative mechanism ensures real-time linkage between resource allocation and decompression events, reduces the idle computing power recovery delay to the millisecond level, and supports the improvement of the force target function value. BRIEF DESCRIPTION OF DRAWINGS

[0032] The accompanying drawings, which are included to provide a further understanding of the application and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0033] Figure 1 A flowchart of a power distribution master station computing power resource adaptive scheduling method provided by an embodiment of the application;

[0034] Figure 2 A flowchart of a computing power resource allocation parameter acquisition method provided by an embodiment of the application;

[0035] Figure 3 A flowchart of a distributed asynchronous cooperative mechanism implementation method provided by an embodiment of the application;

[0036] Figure 4 A structural diagram of a power distribution master station computing power resource adaptive scheduling device provided by an embodiment of the application;

[0037] Figure 5 A structural diagram of an electronic device provided by an embodiment of the application. DETAILED DESCRIPTION

[0038] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and embodiments, so that how the present application applies technical means to solve technical problems and achieves technical effects can be fully understood and implemented. It should be noted that, unless there is a conflict, each embodiment in the present application and each feature in each embodiment can be combined with each other, and the formed technical solutions are within the protection scope of the present application.

[0039] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from here.

[0040] Please refer to Figure 1 The power distribution master station computing resource adaptive scheduling method provided by the present application comprises:

[0041] S101, real-time monitoring of the dynamic change of the amount of each service data to be decompressed according to the preset time section division strategy, and obtaining a computing resource allocation scheme through a preset constraint condition according to the change of the amount of data to be decompressed;

[0042] S102, calculating the decompression time length of each service in the current time section according to the computing resource allocation scheme, obtaining the total decompression time length of each service by accumulating the decompression time length of each time section, and calculating the time delay redundancy rate and the timeout percentage based on the total decompression time length;

[0043] S103, constructing a network support force objective function through the time delay redundancy rate and the timeout percentage according to the Markowitz mean-variance model, and constructing a target function according to the network support force with the goal of maximizing the network support force;

[0044] S104, obtaining optimized computing resource allocation parameters and time section division strategies through the target function, dynamically adjusting the power distribution master station resources according to the optimized computing resource allocation parameters, and triggering time section reset when the service decompression is completed to reallocate idle resources.

[0045] The constraint condition comprises a first constraint condition and a second constraint condition; the first constraint condition is that the total computing resource allocated to each service does not exceed the total computing resource of the power distribution master station; and the second constraint condition is that the computing resource allocated to each service is non-negative and does not exceed its maximum demand value.

[0046] Specifically, in actual work, considering that the compression ratio of differentiated services in the network is not constant, therefore, in the process of service decompression, the power distribution master station can adaptively adjust the resource allocation scheme according to the change of the decompression data volume of each service in real time, in order to improve the resource utilization efficiency and reduce the probability of service decompression timeout, the concept of time section is introduced, and the length between the time when the resource allocation is restarted and the last resource allocation time is defined as a time section. The resource allocation scheme in a given time section i needs to meet the following constraints:

[0047] Σ k∈κ f k,i ≤f all ;

[0048] Wherein, κ is a set of services in the network, f k,i is the power resource allocated by the power distribution master station to service k in time section i, f all is the total amount of power resources of the power distribution master station, and the resource allocation scheme in a time section is fixed.

[0049] In an embodiment of the present application, calculating the decompression time of each service in the current time section according to the resource allocation scheme comprises: calculating the decompression time of each service according to the number of processor cycles required for decompression per bit, the amount of data to be decompressed, a preset minimum constant and a time section constant through the following formula:

[0050]

[0051] In the above formula, ρ k is the number of CPU cycles required for decompression per bit of service k, D k,i is the amount of data to be decompressed of service k at the beginning of time section i, is a minimum constant to prevent the case of unbounded solution when f k,i is 0 in the problem solving process, t i is the length of time section i, t k,i is the decompression time of service k.

[0052] In the above embodiment, resource allocation is only performed on the services that have not completed decompression at the beginning of each time section, so the decompression time of the remaining services in the current time section can be recorded as 0. In addition, in order to prevent errors in calculation, when f k,i = 0, that is, the power resource allocated to the decompressed service k is 0, it means that service k does not decompress in the corresponding time section i, and t k,i= 0. Considering the latency requirement of differentiated services and efficient use of computing resources, and facing multiple time sections with uncertain number and length, adaptive adjustment of power distribution master station computing resource allocation scheme will affect the total decompression time length of each service, so the total decompression time length t k is expressed as follows:

[0053] t k = Σ i∈[1,δ] t k,i ;

[0054] Wherein, δ is the number of time sections. Based on the total decompression time length calculation method, after the power distribution master station completes all decompression work, taking service k as an example, the calculation methods of service k decompression latency redundancy rate θ k and timeout percentage γ k are as follows:

[0055]

[0056] Wherein, τ k is the latency requirement of service k, t k ≤ τ k , which means that service k can complete decompression within its latency requirement without timeout, otherwise it means that service k is timeout. In order to improve the service latency redundancy rate and reduce the service timeout percentage, the number of time sections and the length of each time section need to be considered, so the following constraints need to be met:

[0057] ∑ i∈[1,δ] t i = T;

[0058] 1 ≤ δ ≤ |κ|;

[0059] Wherein, T is the total time length required for the power distribution master station to complete all service decompression, and in order to prevent time section fragmentation, the total number is not more than the total number of service types.

[0060] In an embodiment of the present application, the network support force objective function is constructed by the latency redundancy rate and the timeout percentage according to the Markowitz mean-variance model, which includes:

[0061] The network support force objective function is constructed by the following formula:

[0062]

[0063] In the above formula, E(θ k ) is the mean of latency redundancy rate θ k , E(γ k ) is the mean of service timeout percentage γ k , The network support force is for all services in the network. Specifically, in this embodiment, the time delay redundancy rate and the timeout percentage of all services in the network are comprehensively considered to affect the network support force, referring to the Markowitz mean-variance model, the network support force for all services in the network can be represented as Specifically, as described above.

[0064] Further, the target function is constructed according to the network support force with the goal of maximizing the network support force, which includes:

[0065] The analysis model is constructed by the following formula:

[0066]

[0067] In the above formula, The network support force is for all services in the network; the constraint C1 is the computing resource constraint, that is, the sum of the computing resources allocated by the power distribution master station to each service cannot exceed the total computing resources; the constraint C2 represents that the computing resources allocated to each service are non-negative and cannot be greater than the total computing resources; the constraint C3 is the time section quantity constraint, the number of time sections cannot exceed the total number of service types; the constraint C4 is the time section length constraint, the sum of the lengths of each time section cannot exceed the total decompression time of the power distribution master station; k is the service set; δ is the number of time sections; t i The length of time section i; f all The total amount of computing resources of the power distribution master station; f k The computing frequency allocated to each service.

[0068] In this embodiment, the computing resource is redistributed by setting the time section to improve the network support force, and the optimization problem is constructed for the decompression model containing the service set κ, and the optimization goal is to maximize the network support force. Considering the three types of variable resource allocation scheme f, time section length t and time section number δ, the optimization problem can be represented as above.

[0069] Please refer to Figure 2 As shown, the optimal computing resource allocation parameters are obtained by analyzing the target function, which includes:

[0070] S201, constructing a Lagrange function according to the target function, and obtaining the optimal solution of the target function based on the Lagrange function and the KKT condition;

[0071] S202, obtaining the computing resource allocation parameters by iteratively solving the optimal solution of the target function;

[0072] The Lagrange function includes:

[0073]

[0074] The KKT condition includes:

[0075]

[0076] In the above formula, ε, η k and χ k respectively correspond to the Lagrange multipliers of constraints C1 and C2; is the first derivative of , is the optimal solution of the computing power resources allocated to each service, ε * , and are the optimal solutions of ε, η k and χ k ; is the network support force corresponding to the service k affected by the allocated computing power resources f k .

[0077] Specifically, in actual work, the optimal computing power resource allocation scheme for each service is generated based on the computing power resources given by the power distribution master station within a fixed time section i, and the corresponding optimization problem is as follows:

[0078]

[0079] Wherein, is the network support force affected only by the computing power resource allocation scheme, and the influencing factor is the total amount of computing power resources f all of the power distribution master station, and the optimization problem for the computing frequency f k allocated to each service is strictly concave, so there is a unique extreme point, which satisfies the KKT condition, and the corresponding Lagrange function is as follows:

[0080]

[0081] Wherein, ε, η k and χ k respectively correspond to the Lagrange multipliers of constraints C1 and C2, and the KKT condition is as follows:

[0082]

[0083] Wherein, is the first derivative of , is the optimal solution of the computing power resources allocated to each service, ε * , and are the optimal solutions of ε, η k and χ k . Accordingly, the properties of the optimal solution are summarized as follows:

[0084] 1) When , At the same time And

[0085] 2) When And , At this time

[0086] 3) When , And

[0087] According to the above properties, it can be found that And There is a correlation between It will monotonically decrease with the increase of By iterating * According to the relationship between * , And The optimal solution Specifically, when Satisfy property 1), at this time When Satisfy property 3), at this time When And Satisfy property 2). By iterating * Until Approximate all The optimal computing resource allocation scheme And the maximum network support force under a given time section

[0088] In an embodiment of the present application, the time section division strategy obtained by analyzing the target function comprises: analyzing the target function to obtain the time section division strategy by using a distributed asynchronous cooperative mechanism. Specifically, please refer to Figure 3 As shown in the figure, the distributed asynchronous cooperative mechanism comprises:

[0089] S301 When it is detected that any service is completed decompression, interrupt the current time section;

[0090] S302 Release the computing resource occupied by the decompressed service, and dynamically adjust the priority according to the time delay redundancy rate of the remaining service, and re-allocate the computing resource;

[0091] S303 Based on the updated to-be-decompressed data volume and the target function, a new time section division scheme is generated;

[0092] Wherein the time section division step is dynamically adjusted according to the type of service.

[0093] Specifically, in actual work,

[0094] According to the original problem OP, in combination with the feedback of the computing resource allocation layer, considering the to-be-decompressed data volume of the service, the optimization problem of time section division can be expressed as follows:

[0095]

[0096] Wherein represents the network support force under the given time section division scheme {t, δ}. The network support force obtained by solving the computing resource allocation problem involves a distributed asynchronous collaborative mechanism for solving the above problem facing the time section division. The specific process is shown in Table 1:

[0097] Table 1

[0098]

[0099] Wherein, γ is the time section division step. In step 5 of Table 1, under the given computing resource allocation scheme within the time section, if the service k has completed decompression, the corresponding computing resource cannot be reused within this time section, therefore, in order to improve resource utilization and maximize network support force, the time section needs to be reset and the computing resource needs to be allocated. In this embodiment, the core idea of the distributed asynchronous collaborative mechanism is that the power distribution master station only updates the optimal computing resource allocation scheme when there is a service that has completed decompression, without considering whether the remaining services have completed decompression. The guarantee of decompression delay relies on the driving of network support force. When the time delay of the remaining services is too low, more computing resources will be automatically obtained.

[0100] Please refer to Figure 4As shown, the application also provides a power distribution master station computing resource adaptive scheduling device, which comprises a monitoring module, a calculation module, a construction module and an analysis module; the monitoring module is used for monitoring the dynamic change of the amount of each service to be decompressed in real time according to a preset time section division strategy, and obtaining a computing resource allocation scheme through a preset constraint condition analysis according to the change of the amount of data to be decompressed; the calculation module is used for calculating the decompression time length of each service within the current time section according to the computing resource allocation scheme, obtaining the total decompression time length of each service by accumulating the decompression time length of each time section, and calculating the time delay redundancy rate and the timeout percentage based on the total decompression time length; the construction module is used for constructing a network support force objective function through the time delay redundancy rate and the timeout percentage according to the Markowitz mean-variance model, and constructing an objective function with the maximum network support force as the target according to the network support force; the analysis module is used for obtaining optimized computing resource allocation parameters and time section division strategies through the objective function analysis, dynamically adjusting the power distribution master station resources according to the optimized computing resource allocation parameters, and triggering time section reset when the service decompression is completed to re-allocate idle resources.

[0101] Since the principle of solving problems of the device is similar to the power distribution master station computing resource adaptive scheduling method, the implementation of the device can be referred to the implementation of the power distribution master station computing resource adaptive scheduling method, and the repeated parts will not be described again.

[0102] The beneficial technical effects of the application are that: by adaptively allocating computing resources, the total computing resource utilization rate of the power distribution master station is improved, especially in the data volume fluctuation scenario. The decompression timeout percentage of differentiated services is reduced, and the time delay redundancy rate of key services (such as fault alarm) is improved. The variable step size time section division strategy enables the system to quickly respond to changes in service demand, and the time section division granularity can be dynamically adjusted according to the load, reducing the calculation overhead. The distributed asynchronous cooperative mechanism ensures that resource allocation and decompression events are linked in real time, the idle computing power recovery delay is reduced to milliseconds, and the support force objective function value is improved.

[0103] The application also provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0104] The application also provides a computer readable storage medium storing a computer program for executing the above method.

[0105] The application also provides a computer program product comprising computer programs / instructions, which, when executed by a processor, implement the steps of the above method.

[0106] As Figure 5As shown, the electronic device 600 may also include: a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily need to include these components. Figure 5 All components shown; in addition, the electronic device 600 may also include Figure 5 For components not shown, please refer to existing technologies.

[0107] like Figure 5 As shown, the central processing unit 100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operation of various components of the electronic device 600.

[0108] The memory 140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 100 may execute the program stored in the memory 140 to perform information storage or processing, etc.

[0109] Input unit 120 provides input to central processing unit 100. Input unit 120 may be, for example, a keypad or touch input device. Power supply 170 provides power to electronic device 600. Display 160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0110] The memory 140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142 for storing application programs and function programs or processes for executing the operation of the electronic device 600 via the central processing unit 100.

[0111] The memory 140 may also include a data storage unit (data 143) for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit (driver 144) of the memory 140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0112] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processor 100 to provide input signals and receive output signals, as is the case with conventional mobile communication terminals.

[0113] Based on different communication technologies, a plurality of communication modules 110, such as a cellular network module, a Bluetooth module, and / or a wireless LAN module, can be provided in the same electronic device. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby implementing the usual telecommunication functions. The audio processor 130 can include any suitable buffers, decoders, amplifiers, etc. In addition, the audio processor 130 is coupled to the central processor 100, thereby enabling recording on the local device via the microphone 132 and playing of stored sounds on the local device via the speaker 131.

[0114] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, a system, or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0115] The present application is described in terms of flowcharts and / or block diagrams, which can be implemented by computer program instructions. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing element or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 The flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing element or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 The flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing element or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks.

[0116] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart and / or block diagram block or blocks. Figure 1one or more processes and / or blocks Figure 1 the function specified in the one or more blocks.

[0117] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, so that the instructions executed on the computer or other programmable devices provide processes for implementing the flow Figure 1 one or more processes and / or blocks Figure 1 the function specified in the one or more blocks.

[0118] The specific embodiments described above are intended to be illustrative of the present application, and are not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A power distribution master station computing resource adaptive scheduling method, characterized in that, The method comprises: According to the preset time section division strategy, the dynamic changes of the to-be-decompressed data volume of each service are monitored in real time, and the power resource allocation scheme is obtained through preset constraint condition analysis according to the changes of the to-be-decompressed data volume; According to the power resource allocation scheme, the decompression time length of each service in the current time section is calculated, the total decompression time length of each service is obtained by accumulating the decompression time length of each time section, and the time delay redundancy rate and the timeout percentage are calculated based on the total decompression time length; According to the Markowitz mean-variance model, a network support force objective function is constructed through the time delay redundancy rate and the timeout percentage, and a target function is constructed with the goal of maximizing the network support force according to the network support force; The optimized power resource allocation parameters and time section division strategy are obtained through analysis of the target function, the power resource allocation parameters after optimization are used to dynamically adjust the power distribution master station resources, and the time section is reset when the service decompression is completed to re-allocate idle resources.

2. The power distribution master station computing resource adaptive scheduling method according to claim 1, characterized in that, The constraint condition comprises a first constraint condition and a second constraint condition; The first constraint condition is that the total power resource allocated to each service does not exceed the total power resource of the power distribution master station; and the second constraint condition is that the power resource allocated to each service is non-negative and does not exceed its maximum demand value.

3. The power distribution master station computing resource adaptive scheduling method of claim 1, wherein, According to the power resource allocation scheme, the decompression time length of each service in the current time section is calculated, the total decompression time length of each service is obtained by accumulating the decompression time length of each time section, and the time delay redundancy rate and the timeout percentage are calculated based on the total decompression time length; According to the Markowitz mean-variance model, a network support force objective function is constructed through the time delay redundancy rate and the timeout percentage, and a target function is constructed with the goal of maximizing the network support force according to the network support force; In the above formula, p k Dk is the number of CPU cycles required to decompress one bit of service k, k,i Dk is the number of CPU cycles required to decompress one bit of service k, is a very small constant to prevent the situation where f k,i is 0, t i is the length of time section i, t k,i is the decompression duration of service k.

4. The power distribution master station computing resource adaptive scheduling method of claim 1, wherein, According to the Markowitz mean-variance model, a network support force objective function is constructed through the time delay redundancy rate and the timeout percentage, and a target function is constructed with the goal of maximizing the network support force according to the network support force; The analysis model is constructed through the following formula: In the above equation, E(θ k ) is the mean of the delay redundancy rate θ k , E(γ k ) is the mean of the service timeout percentage γ k , and is the network support force for all services in the network.

5. The power distribution master station computing resource adaptive scheduling method according to claim 1, characterized in that, The optimized power resource allocation parameters are obtained through analysis of the target function, which comprises: A Lagrange function is constructed according to the target function, and the optimal solution of the target function is obtained based on the Lagrange function and the KKT condition; In the above formula, The network support force is for all services in the network; constraint C1 is the computing resource constraint, that is, the sum of the computing resources allocated by the power distribution master station to each service cannot exceed the total computing resources; constraint C2 indicates that the computing resources allocated to each service are non-negative and cannot be greater than the total computing resources; constraint C3 is the time section quantity constraint, the number of time sections cannot exceed the total number of service types; constraint C4 is the time section length constraint, the sum of the lengths of each time section cannot exceed the total service decompression time of the power distribution master station; k is the service set; δ is the number of time sections; t i The length of time section i is t i ; f all The total amount of computing resources of the power distribution master station is f k The computing frequency allocated to each service is f k.

6. The power distribution master station computing resource adaptive scheduling method of claim 5, wherein, The power resource allocation parameters are obtained by iteratively solving the optimal solution of the target function; The Lagrange function comprises: The KKT condition comprises: The time section division strategy is obtained by analyzing the target function using a distributed asynchronous collaborative mechanism. The distributed asynchronous collaborative mechanism comprises: In the above formula, ε, η k and χ k respectively correspond to the Lagrange multipliers of constraints C1 and C2; is the first derivative of , is the optimal solution of the computing power resources allocated to each service, ε * , and are the optimal solutions of ε, η k and χ k ; is the network support force corresponding to the service k affected by the allocated computing power resource f k .

7. The power distribution master station computing resource adaptive scheduling method of claim 5, wherein, When it is detected that any service completes decompression, the current time section is interrupted; 8. The power distribution master station computing resource adaptive scheduling method of claim 7, wherein, The power resource occupied by the decompressed service is released, the priority is dynamically adjusted according to the time delay redundancy rate of the remaining service, and the power resource is re-allocated; Based on the updated to-be-decompressed data volume and the target function, a new time section division scheme is generated; The time section division step is dynamically adjusted according to the type of service. The device comprises a monitoring module, a calculation module, a construction module, and an analysis module; ​ 9. A power distribution master station computing resource adaptive scheduling apparatus, characterized in that, ​ The monitoring module is configured to monitor dynamic changes of the to-be-decompressed data volume of each service in real time according to a preset time section division strategy, and obtain an algorithm resource allocation scheme through preset constraint condition analysis according to the changes of the to-be-decompressed data volume; The computing module is configured to calculate the decompression time length of each service within a current time section according to the algorithm resource allocation scheme, obtain the total decompression time length of each service by accumulating the decompression time lengths of each time section, and calculate the time delay redundancy rate and the timeout percentage based on the total decompression time length; The constructing module is configured to construct a network support force objective function through the time delay redundancy rate and the timeout percentage according to a Markowitz mean-variance model, and construct a target function with the maximum network support force as the target according to the network support force; The analysis module is configured to obtain optimized algorithm resource allocation parameters and time section division strategies through the target function analysis, dynamically adjust the power distribution master station resources according to the optimized algorithm resource allocation parameters, and trigger time section reset when the service decompression is completed, so as to re-allocate idle resources.

10. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the method of any one of claims 1 to 8.

11. A computer readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the method of any one of claims 1 to 8.

12. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to realize the steps of the method of any one of claims 1 to 8.

Images