Edge computing task scheduling method, system and equipment and storage medium

Abstract

The invention discloses an edge computing task scheduling method, system and device and a storage medium, and the method comprises the steps: obtaining a first computing task request type, a target node position, a second computing task request type and a first threshold table, and obtaining a target node position based on the first computing task request type and the second computing task request type; calculating the similarity; when the similarity is greater than a preset threshold value, calculating a global optimal position and an individual optimal position according to the target node position and a first threshold value table, and calculating a distribution node of the first calculation task request at the next moment according to the first calculation task request type, the global optimal position, the individual optimal position and the first threshold value table, thereby improving the resource utilization rate; and the task response time is shortened.

Description

Technical Field

[0001] The present application relates to the technical field related to edge computing task scheduling, and in particular to an edge computing task scheduling method, system, device and storage medium. Background Art

[0002] With the advent of the Internet of Things era, edge computing, as a key area of ​​the new generation of intelligent cloud computing technology, can realize real-time processing of complex intelligent applications at the edge. Large-scale data does not need to experience the delay of cloud transmission, thereby greatly improving user service quality.

[0003] Currently, scheduling edge containers for scientific and technological data resources primarily relies on traditional static or heuristic rule-based methods. For example, Kubernetes, a mainstream container orchestration platform, typically uses a scheduling mechanism based on resource allocation strategies. These include static scheduling based on resource requirements such as CPU, memory, and network bandwidth, as well as dynamic scheduling based on node load. However, these methods suffer from poor adaptability, lack of accurate prediction, and limited optimization capabilities in practical applications. Summary of the Invention

[0004] This application aims to at least solve the technical problems existing in the prior art. To this end, this application proposes an edge computing task scheduling method, system, device and storage medium that can improve resource utilization and shorten task response time.

[0005] In a first aspect of the present application, a method for scheduling edge computing tasks is provided, comprising the following steps:

[0006] Obtain a first computing task request type, a target node location, a second computing task request type, and a first threshold table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node location is the location of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; and the first threshold table is the load value of each node at the current moment;

[0007] Calculating a similarity based on the first computing task request type and the second computing task request type;

[0008] When the similarity is greater than a preset threshold, calculating a global optimal position and an individual optimal position according to the target node position and the first threshold table;

[0009] An allocation node for the first computing task request at the next moment is calculated according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table.

[0010] The control method according to the embodiment of the present application has at least the following beneficial effects:

[0011] This method obtains a first computing task request type, a target node position, a second computing task request type and a first threshold table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node position is the position of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; the first threshold table is the load value of each node at the current moment; based on the first computing task request type and the second computing task request type, similarity is calculated; when the similarity is greater than a preset threshold, the global optimal position and the individual optimal position are calculated according to the target node position and the first threshold table. This application realizes efficient, flexible and low-latency scientific and technological data resource scheduling by calculating the global optimal position and the individual optimal position according to the target node position and the first threshold table, and calculates the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table, thereby improving resource utilization and shortening task response time.

[0012] According to some embodiments of the present application, calculating the global optimal position and the individual optimal position according to the target node position and the first threshold table includes:

[0013] Obtaining a first velocity of the target node at the current moment;

[0014] The target node position is taken as the first individual optimal position; the node with the smallest load value among all nodes is taken as the first global optimal position;

[0015] Calculating a first position according to the target node position, the first speed, the first individual optimal position, and the first global optimal position;

[0016] Allocating the first computing task request type to the node corresponding to the first position; and calculating a second load value of the node corresponding to the first position at a next moment;

[0017] Based on the second load value and the load value of the target node at the current moment, selecting the one with the smaller load value as the individual optimal position;

[0018] The individual optimal positions corresponding to all nodes are calculated, and the one with the smallest load value is selected as the global optimal position.

[0019] According to some embodiments of the present application, the first position is calculated using the following formula:

[0020] ΔT i =Threshold(x i (t))-Ucurrent (x i (t));

[0021] v i (t+1)=wv i (t)+c1r1(p i -x i (t))+c2r2(gx i (t))+ε·ΔT i ;

[0022] x i (t+1)=x i (t)+v i (t+1);

[0023] Where, ΔT i is the difference, Threshold(x i (t)) is the load value of the target node at the current moment in the first threshold table, U current (x i (t)) is the load value of the target node at the current moment, v i (t+1) is the second speed, x i (t+1) is the first position, w is the preset proportional coefficient, v i (t) is the first speed, c1 is the first preset parameter, c2 is the second preset parameter, r1 is the first random factor, r2 is the second random factor, p i is the optimal position of the first individual, x i (t) is the target node position, and ε is the third preset parameter.

[0024] According to some embodiments of the present application, the calculating, based on the first computing task request type, the global optimal position, the individual optimal position, and the first threshold table, an allocation node for the first computing task request at the next moment includes:

[0025] Obtaining the predicted time for the target node to complete the first computing task, the computing resource amount of the first computing task, the QoS load value of the first computing task, and the maximum load value in the first threshold table;

[0026] The priority score of the first computing task request is calculated using the following formula:

[0027]

[0028] Among them, Score is the priority score, α is the preset response time, β is the computational complexity, γ is the preset QoS weight, and Y expected Y is the predicted time for the target node to complete the first computing task, thresholdThe maximum response time allowed by the preset system, C request is the computing resource amount of the first computing task, C threshold is the maximum load value in the first threshold table, QoS score is the QoS load value of the first computing task;

[0029] Calculating a current composite threshold based on the first threshold table;

[0030] When the priority score is less than the current synthesis threshold, allocating the first computing task request to the node corresponding to the global optimal position;

[0031] When the priority score is greater than or equal to the current synthesis threshold, the first computing task request is allocated to the node corresponding to the individual best position.

[0032] According to some embodiments of the present application, calculating the current composite threshold based on the first threshold table includes:

[0033] Get the historical composite threshold of the previous moment and the total system load at the current moment;

[0034] Calculate the average priority score of all nodes at the current moment;

[0035] The current synthetic threshold is calculated based on the historical synthetic threshold, the current total system load, and the priority score average using the following formula:

[0036]

[0037] in, is the current synthesis threshold, Avg(Score histourical ) is the priority score mean, Load current is the total system load at the current moment, is the historical synthesis threshold, and κ is a preset smoothing coefficient.

[0038] According to some embodiments of the present application, the similarity is calculated using the following formula:

[0039]

[0040] Among them, Cosine Similarity is similarity, R i is the type of the actual computing task requested by the target node at the current moment, P i is the type of the i-th predicted computing task request of the target node at the current moment, and n is the total number of actual computing task requests of the target node at the current moment.

[0041] According to some embodiments of the present application, the edge computing task scheduling method further includes:

[0042] When the similarity is less than the preset threshold, based on the first computing task request type, updating the ARIMA prediction model, the ARIMA prediction model being a model that outputs the second computing task request type;

[0043] Allocate the first computing task request to the first node with the smallest load value in the first threshold table;

[0044] The first load value of the first node is calculated by the following formula:

[0045]

[0046] Among them, T i is the first load value of the first node, w j is the weight of the jth actual computing task request type, n is the total number of actual computing task request types, u i,j is the utilization rate of the jth actual computing task request type of the first node;

[0047] The first threshold table is updated according to the first load value.

[0048] In a second aspect of the present application, an edge computing task scheduling system is provided. The edge computing task scheduling system includes:

[0049] A data acquisition module is configured to acquire a first computing task request type, a target node location, a second computing task request type, and a first threshold value table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node location is the location of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; and the first threshold value table is the load value of each node at the current moment;

[0050] a similarity calculation module, configured to calculate a similarity based on the first computing task request type and the second computing task request type;

[0051] An optimal position calculation module, configured to calculate a global optimal position and an individual optimal position according to the target node position and the first threshold table when the similarity is greater than a preset threshold;

[0052] The task request allocation module is used to calculate the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table.

[0053] This system obtains the first computing task request type, the target node position, the second computing task request type and the first threshold table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node position is the position of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; the first threshold table is the load value of each node at the current moment; based on the first computing task request type and the second computing task request type, the similarity is calculated; when the similarity is greater than the preset threshold, the global optimal position and the individual optimal position are calculated according to the target node position and the first threshold table. This application realizes efficient, flexible and low-latency scientific and technological innovation data resource scheduling by calculating the global optimal position and the individual optimal position according to the target node position and the first threshold table, and calculates the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table, thereby improving resource utilization and shortening task response time.

[0054] The third aspect of the present application provides an edge computing task scheduling electronic device, comprising at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the above-mentioned edge computing task scheduling method.

[0055] In a fourth aspect of the present application, a computer-readable storage medium is provided, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned edge computing task scheduling method.

[0056] It should be noted that the beneficial effects between the second to fourth aspects of this application and the prior art are the same as the beneficial effects between the above-mentioned edge computing task scheduling system and the prior art, and will not be described in detail here.

[0057] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become obvious from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0059] Figure 1 This is a flow chart of an edge computing task scheduling method according to an embodiment of the present application;

[0060] Figure 2 This is a schematic diagram of the structure of an embodiment of the edge computing task scheduling system provided by this application;

[0061] Figure 3 It is a structural diagram of an embodiment of the electronic device provided by this application. DETAILED DESCRIPTION

[0062] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and are not to be construed as limiting the present application.

[0063] In the description of this application, if there is a description of first, second, etc., it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0064] In the description of this application, it should be understood that descriptions involving orientation, such as the orientation or positional relationship indicated by up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0065] In the description of this application, it should be noted that, unless otherwise clearly defined, terms such as setting, installing, and connecting should be understood in a broad sense, and technical personnel in the relevant technical field can reasonably determine the specific meaning of the above terms in this application based on the specific content of the technical solution.

[0066] As the amount of information and data increases, users have higher and higher expectations and requirements for the ability to retrieve and extract information from text. For example, users expect to be able to directly obtain the exact related information corresponding to the search term through retrieval or search.

[0067] With the advent of the Internet of Things era, edge computing, as a key area of ​​the new generation of intelligent cloud computing technology, can realize real-time processing of complex intelligent applications at the edge. Large-scale data does not need to experience the delay of cloud transmission, thereby greatly improving user service quality.

[0068] Currently, scheduling edge containers for scientific and technological data resources primarily relies on traditional static or heuristic rule-based methods. For example, Kubernetes, a mainstream container orchestration platform, typically uses a scheduling mechanism based on resource allocation strategies. These include static scheduling based on resource requirements such as CPU, memory, and network bandwidth, as well as dynamic scheduling based on node load. However, these methods suffer from poor adaptability, lack of accurate prediction, and limited optimization capabilities in practical applications.

[0069] In order to solve the above technical defects, the embodiments of the present application provide an edge computing task scheduling method, system, device and storage medium.

[0070] See Figure 1 , is a flow chart of an edge computing task scheduling method provided by an embodiment of the present application, which is applied to an electronic device, which may be a server, etc. Figure 1 As shown, the edge computing task scheduling method includes:

[0071] Step S101: Obtain a first computing task request type, a target node location, a second computing task request type, and a first threshold table, wherein the first computing task request type is the type of the actual computing task request; the target node location is the location of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; and the first threshold table is the load value of each node at the current moment.

[0072] Step S102: Calculate similarity based on the first computing task request type and the second computing task request type;

[0073] Step S103: When the similarity is greater than a preset threshold, the global optimal position and the individual optimal position are calculated according to the target node position and the first threshold table;

[0074] Step S104: Calculate the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table.

[0075] The above nodes are the basic units in the computing network, which can be physical or virtual devices with no specific functional restrictions; container nodes are dedicated nodes for running containerized applications, focusing on the container runtime environment and resource scheduling; edge nodes are computing devices deployed close to data sources or users, performing local processing, low latency, local data processing, and possible resource constraints.

[0076] This method obtains a first computing task request type, a target node position, a second computing task request type and a first threshold table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node position is the position of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; the first threshold table is the load value of each node at the current moment; based on the first computing task request type and the second computing task request type, similarity is calculated; when the similarity is greater than a preset threshold, the global optimal position and the individual optimal position are calculated according to the target node position and the first threshold table. This application realizes efficient, flexible and low-latency scientific and technological data resource scheduling by calculating the global optimal position and the individual optimal position according to the target node position and the first threshold table, and calculates the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table, thereby improving resource utilization and shortening task response time.

[0077] In some embodiments, in step S103, calculating the global optimal position and the individual optimal position according to the target node position and the first threshold table includes:

[0078] Step S201: Obtain the first speed of the target node at the current moment;

[0079] Step S202: The target node position is used as the first individual optimal position; the node with the smallest load value among all nodes is used as the first global optimal position;

[0080] Step S203, calculating a first position according to the target node position, the first speed, the first individual optimal position and the first global optimal position;

[0081] Step S204: Allocate the first computing task request type to the node corresponding to the first position; and calculate a second load value of the node corresponding to the first position at a next moment;

[0082] Step S205: Based on the second load value and the load value of the target node at the current moment, select the one with the smaller load value as the individual optimal position;

[0083] Step S206: Calculate the individual optimal positions corresponding to all nodes, and select the one with the smallest load value as the global optimal position.

[0084] In some embodiments, in step S203, the first position is calculated using the following formula:

[0085] ΔT i =Threshold(x i (t))-U current (x i (t));

[0086] v i (t+1)=wv i (t)+c1r1(p i -x i (t))+c2r2(gx i (t))+ε·ΔT i ;

[0087] x i (t+1)=x i (t)+v i (t+1);

[0088] Where, ΔT i is the difference, Threshold(x i (t)) is the load value of the target node at the current moment in the first threshold table, U current (x i (t)) is the load value of the target node at the current moment, v i (t+1) is the second speed, x i (t+1) is the first position, w is the preset proportional coefficient, v i (t) is the first speed, c1 is the first preset parameter, c2 is the second preset parameter, r1 is the first random factor, r2 is the second random factor, p i is the optimal position of the first individual, x i (t) is the target node position, and ε is the third preset parameter.

[0089] Specifically, in some embodiments, computing node A has a CPU utilization rate of 50% and a memory utilization rate of 60%. Computing node B has a CPU utilization rate of 80% and a memory utilization rate of 75%. Computing node C has a CPU utilization rate of 90% and a memory utilization rate of 85%. Computing nodes D and E have relatively light loads (30% and 40%, respectively). According to the PSO algorithm, tasks are assigned to nodes A, D, and E, avoiding concentrating all tasks on nodes B and C, thereby ensuring a balanced distribution of computing resources.

[0090] This application uses a particle swarm optimization algorithm to dynamically adjust the load of computing nodes during task allocation to achieve optimal resource utilization and shortest task response time.

[0091] In some embodiments, in step S104, calculating the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position, and the first threshold table includes:

[0092] Step S301: Obtain the predicted time for the target node to complete the first computing task, the computing resource amount of the first computing task, the QoS load value of the first computing task, and the maximum load value in the first threshold table;

[0093] Step S302: Calculate the priority score of the first computing task request using the following formula:

[0094]

[0095] Among them, Score is the priority score, α is the preset response time, β is the computational complexity, γ is the preset QoS weight, T expected T is the predicted time for the target node to complete the first computing task, threshold The maximum response time allowed by the preset system, C request is the computing resource amount of the first computing task, C threshold is the maximum load value in the first threshold table, QoS score is the QoS load value of the first computing task;

[0096] Step S303: Calculate the current synthesis threshold based on the first threshold table;

[0097] Step S304: When the priority score is less than the current synthesis threshold, the first computing task request is assigned to the node corresponding to the global optimal position;

[0098] Step S305: When the priority score is greater than or equal to the current synthesis threshold, the first computing task request is allocated to the node corresponding to the individual best position.

[0099] This application combines threshold tables for resource management, optimizes the matching of computing resources and task requirements through real-time load modeling, and improves the intelligence of scheduling.

[0100] In some embodiments, in step S303, calculating the current composite threshold based on the first threshold table includes:

[0101] Step S401: Obtain the historical composite threshold at the previous moment and the total system load at the current moment;

[0102] Step S402: Calculate the average priority score of all nodes at the current moment;

[0103] Step S403: Calculate the current composite threshold based on the historical composite threshold, the current system total load, and the priority score average using the following formula:

[0104]

[0105] in, is the current synthesis threshold, Avg(Scorehistourical ) is the mean priority score, Load current is the total system load at the current moment, is the historical synthesis threshold, and κ is the preset smoothing coefficient.

[0106] This application improves the accuracy of data by updating the synthesis threshold in real time.

[0107] In some embodiments, in step S102, the similarity is calculated using the following formula:

[0108]

[0109] Among them, Cosine Similarity is similarity, R i is the type of the actual computing task requested by the target node at the current moment, P i is the type of the i-th predicted computing task request of the target node at the current moment, and n is the total number of actual computing task requests of the target node at the current moment.

[0110] In some embodiments, the edge computing task scheduling method further includes:

[0111] Step S501: When the similarity is less than a preset threshold, the ARIMA prediction model is updated based on the first computing task request type, and the ARIMA prediction model is a model that outputs the second computing task request type;

[0112] Step S502: Allocate the first computing task request to the first node with the smallest load value in the first threshold table;

[0113] Step S503: Calculate the first load value of the first node using the following formula:

[0114]

[0115] Among them, T i is the first load value of the first node, w j is the weight of the jth actual computing task request type, n is the total number of actual computing task request types, u i,j is the utilization rate of the jth actual computing task request type of the first node;

[0116] Step S504: Update the first threshold table according to the first load value.

[0117] This application introduces time series prediction to achieve task request prediction based on historical data, thereby improving the timeliness of task scheduling.

[0118] In this application, edge computing is to distribute computing tasks from data centers to edge nodes that are closer to data sources or end users to reduce latency and bandwidth consumption.

[0119] The container technology in this application is a lightweight virtualization technology that isolates different applications and services through containers to achieve rapid deployment, portability and scalability.

[0120] The ARIMA forecasting model in this application is a time series forecasting model that describes and predicts time series data through autoregression, differencing, and moving average.

[0121] The threshold table in this application is a data table that stores preset thresholds and corresponding optimization strategies to help dynamically adjust the parameters of the algorithm to adapt to load changes.

[0122] Specifically, to facilitate understanding by those skilled in the art, a set of best embodiments are provided below:

[0123] 1. Data Acquisition

[0124] Obtain a first computing task request type, a target node location, a second computing task request type, and a first threshold table, wherein the first computing task request type is the type of the actual computing task request; the target node location is the location of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; and the first threshold table is the load value of each node at the current moment;

[0125] 2. Similarity calculation:

[0126] Based on the first computing task request type and the second computing task request type, the similarity is calculated using the following formula:

[0127]

[0128] Among them, Cosine Similarity is similarity, R i is the type of the actual computing task requested by the target node at the current moment, P i is the type of the i-th predicted computing task request of the target node at the current moment, and n is the total number of actual computing task requests of the target node at the current moment.

[0129] 3. Calculation of global optimal position and individual optimal position:

[0130] When the similarity is greater than the preset threshold, the global optimal position and the individual optimal position are calculated according to the target node position and the first threshold table, specifically:

[0131] Get the first velocity of the target node at the current moment;

[0132] The target node position is taken as the first individual optimal position; the node with the smallest load value among all nodes is taken as the first global optimal position;

[0133] The first position is calculated according to the target node position, the first velocity, the first individual optimal position, and the first global optimal position using the following formula:

[0134] ΔT i =Threshold(x i (t))-U current (x i (t));

[0135] v i (t+1)=wv i (t)+c1r1(p i -x i (t))+c2r2(gx i (t))+ε·ΔT i ;

[0136] x i (t+1)=x i (t)+v i (t+1);

[0137] Where, ΔT i is the difference, Threshold(x i (t)) is the load value of the target node at the current moment in the first threshold table, U current (x i (t)) is the load value of the target node at the current moment, v i (t+1) is the second speed, x i (t+1) is the first position, w is the preset proportional coefficient, v i (t) is the first speed, c1 is the first preset parameter, c2 is the second preset parameter, r1 is the first random factor, r2 is the second random factor, p i is the optimal position of the first individual, x i (t) is the target node position, and ε is the third preset parameter.

[0138] Allocating the first computing task request type to the node corresponding to the first position; and calculating a second load value of the node corresponding to the first position at a next moment;

[0139] Based on the second load value and the load value of the target node at the current moment, the node with the smaller load value is selected as the individual optimal position;

[0140] Calculate the individual optimal positions corresponding to all nodes, and select the one with the smallest load value as the global optimal position.

[0141] 4. Task Request Allocation:

[0142] The allocation node of the first computing task request at the next moment is calculated according to the first computing task request type, the global optimal position, the individual optimal position, and the first threshold table, specifically:

[0143] Obtaining the predicted time for the target node to complete the first computing task, the computing resource amount of the first computing task, the QoS load value of the first computing task, and the maximum load value in the first threshold table;

[0144] The priority score of the first computing task request is calculated using the following formula:

[0145]

[0146] Among them, Score is the priority score, α is the preset response time, β is the computational complexity, γ is the preset QoS weight, T expected T is the predicted time for the target node to complete the first computing task, threshold The maximum response time allowed by the preset system, C request is the computing resource amount of the first computing task, C threshold is the maximum load value in the first threshold table, QoS score is the QoS load value of the first computing task;

[0147] The current synthetic threshold is calculated based on the first threshold table, specifically:

[0148] Get the historical composite threshold of the previous moment and the total system load at the current moment;

[0149] Calculate the average priority score of all nodes at the current moment;

[0150] The current synthetic threshold is calculated based on the historical synthetic threshold, the total system load at the current moment, and the average priority score using the following formula:

[0151]

[0152] in, is the current synthesis threshold, Avg(Score histourical ) is the mean priority score, Load current is the total system load at the current moment, is the historical synthesis threshold, and k is the preset smoothing coefficient.

[0153] When the priority score is less than the current synthesis threshold, the first computing task request is assigned to the node corresponding to the global best position;

[0154] When the priority score is greater than or equal to the current synthesis threshold, the first computing task request is assigned to the node corresponding to the individual's best position.

[0155] 5. Model update and threshold table update when mismatch occurs:

[0156] When the similarity is less than a preset threshold, the ARIMA prediction model is updated based on the first computing task request type, and the ARIMA prediction model is a model that outputs the second computing task request type;

[0157] Allocate the first computing task request to the first node with the smallest load value in the first threshold table;

[0158] The first load value of the first node is calculated using the following formula:

[0159]

[0160] Among them, T i is the first load value of the first node, w j is the weight of the jth actual computing task request type, n is the total number of actual computing task request types, u i,j is the utilization rate of the jth actual computing task request type of the first node;

[0161] The first threshold table is updated according to the first load value.

[0162] This application is based on time series analysis and particle swarm optimization algorithm, and realizes adaptive scheduling and intelligent optimization of edge containers for scientific and technological data resources through the combination of prediction, monitoring and optimization. First, the access pattern of scientific and technological data resources is modeled using time series analysis models such as ARIMA to predict the type, quantity and computing requirements of future requests, so as to optimize the container resource configuration in advance. Then, the load of edge nodes is monitored in real time, and a threshold table is set. When the computing resource occupancy or data processing load exceeds the set threshold, container migration or resource adjustment is triggered to ensure the efficiency of data calculation and storage. Then, the particle swarm optimization algorithm is used to optimize multiple scheduling objectives (such as load balancing, data processing efficiency, resource utilization and scheduling delay) so that the container can be optimally scheduled according to the dynamic needs of scientific and technological data.

[0163] In addition, refer to Figure 2 One embodiment of the present application provides an edge computing task scheduling system, including a data acquisition module 1100, a similarity calculation module 1200, an optimal position calculation module 1300, and a task request allocation module 1400, wherein:

[0164] The data acquisition module 1100 is used to obtain a first computing task request type, a target node location, a second computing task request type, and a first threshold value table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node location is the location of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; and the first threshold value table is the load value of each node at the current moment.

[0165] The similarity calculation module 1200 is used to calculate the similarity based on the first computing task request type and the second computing task request type;

[0166] The optimal position calculation module 1300 is used to calculate the global optimal position and the individual optimal position according to the target node position and the first threshold table when the similarity is greater than a preset threshold;

[0167] The task request allocation module 1400 is used to calculate the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table.

[0168] This system obtains the first computing task request type, the target node position, the second computing task request type and the first threshold table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node position is the position of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; the first threshold table is the load value of each node at the current moment; based on the first computing task request type and the second computing task request type, the similarity is calculated; when the similarity is greater than the preset threshold, the global optimal position and the individual optimal position are calculated according to the target node position and the first threshold table. This application realizes efficient, flexible and low-latency scientific and technological innovation data resource scheduling by calculating the global optimal position and the individual optimal position according to the target node position and the first threshold table, and calculates the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table, thereby improving resource utilization and shortening task response time.

[0169] It should be noted that this system embodiment and the above-mentioned method embodiment are based on the same inventive concept, so the relevant content of the above-mentioned method embodiment is also applicable to this system embodiment and will not be repeated here.

[0170] Figure 3 A schematic diagram of the hardware structure of edge computing task scheduling provided in an embodiment of the present application is shown.

[0171] The edge computing task scheduling device may include a processor 301 and a memory 302 storing computer program instructions.

[0172] Specifically, the processor 301 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0173] The memory 302 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 302 may include removable or non-removable (or fixed) media. Where appropriate, the memory 302 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 302 is a non-volatile solid-state memory.

[0174] In some embodiments, the memory 302 may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0175] The processor 301 implements any one of the edge computing task scheduling methods in the above embodiments by reading and executing computer program instructions stored in the memory 302.

[0176] In one example, the edge computing task scheduling device may further include a communication interface 303 and a bus 310. Figure 3 As shown, the processor 301 , the memory 302 , and the communication interface 303 are connected via a bus 310 and communicate with each other.

[0177] The communication interface 303 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0178] The bus 310 includes hardware, software or both, and couples the components of the edge computing task scheduling device to each other. For example, and not limitation, the bus may include an accelerated graphics port (AGP) or other graphics bus, an enhanced industry standard architecture (EISA) bus, a front-side bus (FSB), a hypertransport (HT) interconnect, an industry standard architecture (ISA) bus, an infinite bandwidth interconnect, a low pin count (LPC) bus, a memory bus, a microchannel architecture (MCA) bus, a peripheral component interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a serial advanced technology attachment (SATA) bus, a video electronics standard association local (VLB) bus or other suitable bus or a combination of two or more of these. Where appropriate, the bus 310 may include one or more buses. Although the present application describes and illustrates a specific bus, the present application considers any suitable bus or interconnect.

[0179] The edge computing task scheduling device can execute the edge computing task scheduling method in the embodiment of the present application based on the three-dimensional design model, thereby realizing the combination of Figure 1 and Figure 2 Described edge computing task scheduling method and system.

[0180] In addition, in conjunction with the edge computing task scheduling method in the above embodiments, embodiments of the present application may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any of the edge computing task scheduling methods in the above embodiments is implemented.

[0181] It should be understood that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated. Those skilled in the art can make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present application.

[0182] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. Programs or code segments can be stored in machine-readable media, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable media" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0183] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0184] Aspects of the present disclosure have been described above with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in 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, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit. It is also understood that each box in the block diagram and / or flowchart and the combination of the boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0185] The above description is only a specific embodiment of the present application. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the scope of protection of the present application.

Claims

1. A method for scheduling edge computing tasks, characterized in that: The edge computing task scheduling method includes: Obtain a first computing task request type, a target node location, a second computing task request type, and a first threshold table, wherein the first computing task request type is the type of the actual computing task request; the target node location is the location of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; and the first threshold table is the load value of each node at the current moment; Calculating a similarity based on the first computing task request type and the second computing task request type; When the similarity is greater than a preset threshold, calculating a global optimal position and an individual optimal position according to the target node position and the first threshold table; An allocation node for the first computing task request at a next moment is calculated according to the first computing task request type, the global optimal position, the individual optimal position, and the first threshold table.

2. The edge computing task scheduling method according to claim 1, characterized in that: The calculating the global optimal position and the individual optimal position according to the target node position and the first threshold table includes: Obtaining a first velocity of the target node at the current moment; The target node position is taken as the first individual optimal position; the node with the smallest load value among all nodes is taken as the first global optimal position; Calculating a first position according to the target node position, the first speed, the first individual optimal position, and the first global optimal position; Allocating the first computing task request type to the node corresponding to the first position; and calculating a second load value of the node corresponding to the first position at a next moment; Based on the second load value and the load value of the target node at the current moment, selecting the one with the smaller load value as the individual optimal position; The individual optimal positions corresponding to all nodes are calculated, and the one with the smallest load value is selected as the global optimal position.

3. The edge computing task scheduling method according to claim 2, characterized in that: The first position is calculated by the following formula: ΔT i =Threshold(x i (t))-U current (x i (t)); v i (t+1)=wv i (t)+c1r1(p i -x i (t))+c2r2(g-x i (t))+ε·ΔT i ; x i (t+1)=x i (t)+v i (t+1); Where, ΔT i is the difference, Threshold(x i (t)) is the load value of the target node at the current moment in the first threshold table, U current (x i (t)) is the load value of the target node at the current moment, v i (t+1) is the second speed, x i (t+1) is the first position, w is the preset proportional coefficient, v i (t) is the first speed, c1 is the first preset parameter, c2 is the second preset parameter, r1 is the first random factor, r2 is the second random factor, p i is the optimal position of the first individual, x i (t) is the target node position, and ε is the third preset parameter.

4. The edge computing task scheduling method according to claim 3, characterized in that: The calculating, according to the first computing task request type, the global optimal position, the individual optimal position, and the first threshold table, an allocation node for the first computing task request at the next moment includes: Obtaining the predicted time for the target node to complete the first computing task, the computing resource amount of the first computing task, the QoS load value of the first computing task, and the maximum load value in the first threshold table; The priority score of the first computing task request is calculated using the following formula: Among them, Score is the priority score, α is the preset response time, β is the computational complexity, γ is the preset QoS weight, T expected is the predicted time for the target node to complete the first computing task, T threshold The maximum response time allowed by the preset system, C request is the computing resource amount of the first computing task, C threshold is the maximum load value in the first threshold table, QoS score is the QoS load value of the first computing task; Calculating a current composite threshold based on the first threshold table; When the priority score is less than the current synthesis threshold, allocating the first computing task request to the node corresponding to the global optimal position; When the priority score is greater than or equal to the current synthesis threshold, the first computing task request is allocated to the node corresponding to the individual best position.

5. The edge computing task scheduling method according to claim 4, characterized in that: The calculating the current composite threshold based on the first threshold table includes: Get the historical composite threshold of the previous moment and the total system load at the current moment; Calculate the average priority score of all nodes at the current moment; The current synthetic threshold is calculated based on the historical synthetic threshold, the current total system load, and the priority score average using the following formula: in, is the current synthesis threshold, Avg(Scor histourical ) is the priority score mean, Load current is the total system load at the current moment, is the historical synthesis threshold, and κ is a preset smoothing coefficient.

6. The edge computing task scheduling method according to claim 1, characterized in that: The similarity is calculated using the following formula: Among them, Cosine Similarity is similarity, R i is the type of the actual computing task requested by the target node at the current moment, P i is the type of the i-th predicted computing task request of the target node at the current moment, and n is the total number of actual computing task requests of the target node at the current moment.

7. The edge computing task scheduling method according to claim 1, characterized in that: The edge computing task scheduling method further includes: When the similarity is less than the preset threshold, based on the first computing task request type, updating the ARIMA prediction model, the ARIMA prediction model being a model that outputs the second computing task request type; Allocate the first computing task request to the first node with the smallest load value in the first threshold table; The first load value of the first node is calculated by the following formula: Among them, T i is the first load value of the first node, w j is the weight of the jth actual computing task request type, n is the total number of actual computing task request types, u i,j is the utilization rate of the jth actual computing task request type of the first node; The first threshold table is updated according to the first load value.

8. An edge computing task scheduling system, characterized in that: The edge computing task scheduling system includes: A data acquisition module is configured to acquire a first computing task request type, a target node location, a second computing task request type, and a first threshold value table, wherein the first computing task request type is the type corresponding to the actual computing task request; the target node location is the location of the node receiving the first computing task; the second computing task request type is the type of the predicted computing task request of the target node at the current moment; and the first threshold value table is the load value of each node at the current moment; a similarity calculation module, configured to calculate a similarity based on the first computing task request type and the second computing task request type; An optimal position calculation module, configured to calculate a global optimal position and an individual optimal position according to the target node position and the first threshold table when the similarity is greater than a preset threshold; The task request allocation module is used to calculate the allocation node of the first computing task request at the next moment according to the first computing task request type, the global optimal position, the individual optimal position and the first threshold table.

9. An edge computing task scheduling device, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute an edge computing task scheduling method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute an edge computing task scheduling method as described in any one of claims 1 to 7.

Images