Edge intelligent distributed data processing method, system and device and storage medium

Abstract

The invention discloses an edge intelligent distributed data processing method, and relates to the field of data processing. The method comprises the following steps: acquiring state data of at least one first edge device; determining at least one data processing task according to the target demand, wherein the data processing task comprises a task type, a processing deadline and a resource demand of the data processing task; based on the task type, the processing deadline and the resource demand of each data processing task, determining the task importance of the corresponding data processing task; according to the importance degree of each task, the computing resource occupancy rate of each first edge device and the network bandwidth occupancy rate, respectively calculating the processing adaptation degree of each first edge device and each data processing task; determining the stability of each first edge device based on the historical failure rate and the operation duration of the first edge device; and determining a second edge device based on the stability and the processing adaptation degree of each first edge device, and processing the data processing task through the second edge device. The data processing efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to an edge intelligent distributed data processing method, system, device and storage medium. BACKGROUND

[0002] With the rapid development of Internet of Things technology, edge computing has gradually become an important technical means in the field of data processing. Edge computing can effectively reduce data transmission delay, improve real-time performance, and relieve the pressure on cloud computing centers by dispersing data processing tasks to edge devices close to data sources. In the fields of intelligent manufacturing, intelligent transportation and industrial Internet of Things, distributed data processing technology has become an important tool to improve system efficiency and reliability.

[0003] Currently, distributed data processing usually monitors the resource usage of edge devices, and allocates data processing tasks to devices with low load to achieve load balancing of the system.

[0004] However, in actual complex edge computing scenarios, distributed data processing tasks have diverse resource requirements, and relying solely on the computing resource status of edge devices for task allocation can easily result in tasks not being allocated to suitable devices for processing, leading to low data processing efficiency. SUMMARY

[0005] The present application provides an edge intelligent distributed data processing method, system, device and storage medium, which solves the problem of low data processing efficiency caused by relying solely on the computing resource status of edge devices for task allocation in the prior art.

[0006] In a first aspect of the present application, an edge intelligent distributed data processing method is provided, comprising: obtaining state data of at least one first edge device, the state data comprising the computing resource occupancy rate, network bandwidth occupancy rate, historical failure rate and running time length of the first edge device; determining at least one data processing task according to a target requirement, the data processing task comprising the task type, processing deadline and resource requirement of the data processing task; determining the task importance of the corresponding data processing task based on the task type, processing deadline and resource requirement of each data processing task; calculating the processing adaptation degree of each first edge device and each data processing task based on the task importance, the computing resource occupancy rate and the network bandwidth occupancy rate of each first edge device; determining the stability of each first edge device based on the historical failure rate and the running time length of the first edge device; determining a second edge device based on the stability and the processing adaptation degree of each first edge device, and performing data processing task processing through the second edge device.

[0007] By adopting the above technical solution, and by acquiring status data such as the computing resource utilization rate, network bandwidth utilization rate, historical failure rate, and runtime of edge devices, and combining this data processing task type, processing period, and resource requirements to determine task importance, a task evaluation mechanism based on multi-dimensional features is established. Then, based on task importance and device resource status, the processing suitability is calculated so that tasks can be assigned to the devices with the highest resource matching degree. At the same time, the device stability is evaluated by historical failure rate and runtime to ensure that tasks are assigned to edge devices with higher reliability. This achieves optimal matching between data processing tasks and device resources, thereby improving data processing efficiency.

[0008] Optionally, when determining the processing suitability between each first edge device and each data processing task based on task importance, the computing resource utilization rate of the first edge device, and the network bandwidth utilization rate, the specific steps include: assigning a priority weight to each data processing task based on the task importance; calculating the computing resource availability rate of the first edge device using the computing resource utilization rate, wherein the computing resource availability rate is the ratio of unused computing resources to the total computing resources; calculating the network transmission availability rate of the first edge device using the network bandwidth utilization rate, wherein the network transmission availability rate is the ratio of unused bandwidth resources to the total bandwidth resources; and determining the processing suitability between each first edge device and each data processing task based on the priority weight and the computing resource availability rate and network transmission availability rate of the edge device.

[0009] By adopting the above technical solution, priority weights are assigned to each data processing task based on task importance. Combined with the availability of computing resources and network transmission of edge devices, a quantitative analysis of task priority and device resource status is achieved. Through the assessment of the availability of computing and network resources, the actual processing capacity of edge devices is accurately reflected. Furthermore, based on the comprehensive calculation of priority weights and resource availability, a scientific and reasonable processing fit is obtained, providing a reliable decision basis for task allocation. By calculating the ratio of unused computing resources to the total computing resources, and the ratio of unused bandwidth resources to the total bandwidth resources, an accurate assessment of the processing capacity of edge devices is achieved. The method of combining task priority weights with resource availability to calculate processing fit not only considers the importance of tasks but also makes full use of the available resources of the devices, improving the resource utilization efficiency of data processing.

[0010] Optionally, determining the processing suitability between each first edge device and each data processing task based on the priority weight and the computing resource availability and network transmission availability of the first edge device specifically includes: calculating the ratio of the computing resource availability of each first edge device to the resource requirement of each data processing task to obtain the computing resource matching degree; calculating the ratio of the network transmission availability of each first edge device to the data transmission requirement of the data processing task to obtain the network resource matching degree; weighting and summing the computing resource matching degree and the network resource matching degree according to preset computing resource matching degree weights and network resource matching degree weights to obtain the total resource matching degree; and calculating the product of the total resource matching degree and the task priority weight to obtain the processing suitability between each first edge device and each data processing task.

[0011] By adopting the above technical solution, the matching degree of computing resources and the matching degree of network resources are obtained by calculating the ratio of the availability of computing resources of edge devices to the resource requirements of data processing tasks, and the ratio of the availability of network transmission to the requirements of data transmission. This realizes the evaluation of the matching degree between edge devices and data processing tasks in different resource dimensions. Furthermore, the matching degree of computing resources and the matching degree of network resources are weighted and summed according to preset weights to obtain the total matching degree of resources. This weighted calculation method can flexibly adjust the importance of different resource dimensions according to actual application scenarios. Finally, the total matching degree of resources is multiplied by the task priority weight to obtain the final processing adaptability. This achieves the coordination and unity of reasonable resource utilization and task priority, and provides a more comprehensive and accurate evaluation basis for task allocation of edge devices.

[0012] Optionally, assigning priority weights to each data processing task based on the task importance specifically includes: dividing the importance of the data processing task into multiple level intervals, each interval corresponding to a different initial weight; calculating the difference between the processing deadline of the data processing task and the current time to determine the time urgency; determining the request priority based on the request frequency of the data processing task within a preset time period; and performing a weighted calculation of the initial weight, the time urgency, and the request priority to obtain the priority weight of the data processing task.

[0013] By adopting the above technical solution, a basic quantitative standard for task importance is established by dividing task importance into multiple level intervals and corresponding to different initial weights. The time urgency is determined by calculating the difference between the processing deadline and the current time, reflecting the timeliness requirements of the task. At the same time, the request priority is determined based on the request frequency within a preset time period, reflecting the actual demand of the task. Finally, the initial weight, time urgency, and request priority are weighted to obtain the priority weight, realizing a multi-dimensional comprehensive evaluation of task priority. This comprehensively reflects the impact of task importance, time urgency, and request frequency, providing a more comprehensive and scientific evaluation mechanism for task priority division, and effectively ensuring that important and urgent tasks can be processed first.

[0014] Optionally, determining the task importance of each data processing task based on its type, processing period, and resource requirements specifically includes: determining a basic sequence number for each data processing task according to a preset task type level table and the task type of each data processing task, wherein the task type level table includes different basic sequence numbers corresponding to different task types; sorting each data processing task by processing period from largest to smallest to determine a time sequence number for each data processing task, with the earlier the task is sorted, the smaller the corresponding time sequence number; sorting each data processing task by resource requirements from smallest to largest to determine a resource sequence number for each data processing task, with the earlier the task is sorted, the smaller the corresponding resource sequence number; and calculating the product of the basic sequence number, the time sequence number, and the resource sequence number to obtain the task importance.

[0015] By adopting the above technical solution, a basic sequence number standard corresponding to different task types is established through a preset task type level table. The time sequence number and resource sequence number are obtained by sorting the tasks based on two dimensions: processing deadline and resource requirements. The final task importance is obtained by calculating the product of the sequence numbers. This achieves a comprehensive evaluation of the importance of task types, time requirements, and resource consumption. The sorting rules of the time sequence number and resource sequence number respectively reflect the characteristics of tasks with ample deadlines and resource conservation being easier to execute. This provides a reasonable importance evaluation mechanism for task scheduling and effectively balances the influence of task type, time, and resources.

[0016] Optionally, determining the stability of each first edge device based on the historical failure rate and the runtime of the first device specifically includes: calculating the ratio of the historical failure rate to a historical failure rate threshold, and using the reciprocal of the ratio as the first stability value; calculating the ratio of the runtime to a runtime threshold, and using the ratio as the second stability value; and weighting and summing the first stability value and the second stability value with corresponding preset weights to obtain the stability of each first edge device.

[0017] By adopting the above technical solution, the first stability value is obtained by calculating the reciprocal of the ratio of the historical failure rate to its threshold, and the second stability value is obtained by calculating the ratio of the running time to its threshold. These values ​​reflect the reliability and usage level of the equipment, respectively. The two stability values ​​are combined by weighted summation to obtain the final equipment stability assessment result. The equipment performance is evaluated from two dimensions: historical operating conditions and equipment usage time. This provides a reliable reference for equipment status for task allocation and effectively reduces the risk of equipment failure during task execution.

[0018] Optionally, determining the second edge device based on the stability of each of the first edge devices and the processing adaptability specifically includes: filtering first edge devices whose stability is greater than the stability threshold and whose processing adaptability is greater than the processing adaptability threshold to obtain candidate edge devices; weighting and summing the stability and the processing adaptability with corresponding preset weights to obtain a comprehensive score for each of the candidate edge devices; and determining the candidate edge device with the highest comprehensive score as the second edge device.

[0019] By adopting the above technical solution, a set of candidate devices is obtained by initially screening edge devices by setting stability thresholds and processing adaptability thresholds, ensuring that the candidate devices have basic stability and processing capabilities. Furthermore, a weighted summation method is used to comprehensively evaluate the device stability and processing adaptability to obtain a comprehensive score of the candidate devices. The candidate device with the highest comprehensive score is selected as the final execution device, realizing the unity of device reliability and task processing efficiency, and providing a scientific device selection method for task allocation in edge computing environments.

[0020] A second aspect of this application provides an edge intelligence distributed data processing system, comprising: The data acquisition module is used to acquire status data of at least one first edge device, the status data including the computing resource utilization rate, network bandwidth utilization rate, historical failure rate and runtime of the first edge device; The task determination module is used to determine at least one data processing task based on target requirements. The data processing task includes the task type, processing period and resource requirements of the data processing task. The importance assessment module is used to determine the importance of each data processing task based on its task type, processing deadline, and resource requirements. The adaptability calculation module is used to calculate the processing adaptability between each first edge device and each data processing task based on the importance of each task, the computing resource utilization rate of each first edge device, and the network bandwidth utilization rate. A stability assessment module is used to determine the stability of each of the first edge devices based on the historical failure rate and the runtime of the first edge devices. The device selection module is used to determine a second edge device based on the stability of each of the first edge devices and the processing adaptability, and to process the data processing task through the second edge device.

[0021] By adopting the above technical solution, and by acquiring status data such as the computing resource utilization rate, network bandwidth utilization rate, historical failure rate, and runtime of edge devices, and combining this data processing task type, processing period, and resource requirements to determine task importance, a task evaluation mechanism based on multi-dimensional features is established. Then, based on task importance and device resource status, the processing suitability is calculated so that tasks can be assigned to the devices with the highest resource matching degree. At the same time, the device stability is evaluated by historical failure rate and runtime to ensure that tasks are assigned to edge devices with higher reliability. This achieves optimal matching between data processing tasks and device resources, thereby improving data processing efficiency.

[0022] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any of the foregoing.

[0023] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed, perform the method described in any of the preceding descriptions. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the architecture of a distributed data processing system for edge intelligence disclosed in an embodiment of this application; Figure 2 This is a flowchart illustrating a distributed data processing method for edge intelligence disclosed in an embodiment of this application; Figure 3 yes Figure 2 A flowchart illustrating a sub-step of step S104; Figure 4 yes Figure 3 A flowchart illustrating a sub-step of step S201; Figure 5 yes Figure 3 A flowchart illustrating a sub-step of step S204; Figure 6 yes Figure 2 A flowchart illustrating a sub-step of step S103; Figure 7 yes Figure 2 A flowchart illustrating a sub-step of step S105; Figure 8 yes Figure 2 Flowchart of a sub-step in step S106 Figure 9 This is a schematic diagram of a distributed data processing system for edge intelligence provided in an embodiment of this application; Figure 10 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application.

[0025] Explanation of reference numerals in the attached diagram: 21. Data acquisition module; 22. Task determination module; 23. Importance assessment module; 24. Adaptability calculation module; 25. Stability assessment module; 26. Device selection module; 901. Processor; 902. Communication bus; 903. User interface; 904. Network interface; 905. Memory. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0027] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0028] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0029] Figure 1 An exemplary system architecture 10 of an edge-intelligent distributed data processing system is shown.

[0030] likeFigure 1 As shown, system architecture 10 may include electronic devices 11, network 12, and edge devices 13. Network 12 serves as the medium for providing a communication link between electronic devices 11 and edge devices 13. Network 12 may include various connection types, such as wired, wireless communication links, or fiber optic cables.

[0031] Users can use electronic device 11 to interact with edge device 13 via network 12 to allocate and execute data processing tasks. Various data processing applications, such as data analysis applications and computing task processing applications, can be installed on electronic device 11.

[0032] Electronic device 11 is hardware, which can be various devices with data processing capabilities, including but not limited to servers, industrial control computers, data processing terminals, and smart gateways.

[0033] Edge device 13 can be an edge node that provides various data processing services, such as an edge computing device that performs calculations and analyses on the data processing tasks assigned by electronic device 11. The edge computing device can perform resource scheduling and task processing on the received data processing tasks and can feed back the processing results to the electronic device.

[0034] The following detailed explanation uses the electronic device side as an example.

[0035] This embodiment discloses a distributed data processing method for edge intelligence. Figure 2 This is a flowchart illustrating a distributed data processing method for edge intelligence disclosed in an embodiment of this application, as shown below. Figure 2 As shown, steps S101 to S106 are included, and the steps are as follows: S101: Obtain status data of at least one first edge device, including the computing resource utilization rate, network bandwidth utilization rate, historical failure rate, and runtime of the first edge device.

[0036] In this embodiment of the application, the first edge device refers to a local data processing unit deployed in a distributed system, such as an edge server, an edge gateway, an industrial controller, etc., used to perform various data processing tasks.

[0037] Status data refers to a set of parameters reflecting the working status of the first edge device during operation, including indicators such as computing resource utilization, network bandwidth utilization, historical failure rate, and runtime. This status data is used to evaluate the processing capability and operational stability of the edge device.

[0038] Specifically, the electronic device obtains the status data of each first edge device through its built-in monitoring module. The monitoring module receives performance monitoring data from each first edge device, including a resource monitoring unit for statistical analysis of CPU and memory usage, a bandwidth monitoring unit for collecting network communication data, a fault monitoring unit for recording abnormal operation of the first edge device, and an operation monitoring unit for accumulating the working time of the first edge device.

[0039] For example, in a distributed system, the CPU utilization of an edge server is 70% and the memory utilization is 50%, resulting in a computing resource utilization rate of (70%+50%) / 2=60%. The uplink bandwidth utilization is 60% and the downlink bandwidth utilization is 40%, resulting in a network bandwidth utilization rate of (60%+40%) / 2=50%. There has been one failure in the past 30 days, resulting in a historical failure rate of 1 / 30=0.033. The cumulative runtime is 360 hours.

[0040] S102: Determine at least one data processing task based on the target requirements. The data processing task includes the task type, processing time, and resource requirements of the data processing task.

[0041] In this embodiment of the application, the data processing task refers to the specific computing task that needs to be performed by the edge device in the distributed system, such as data analysis, image processing, real-time monitoring, etc.

[0042] The task type reflects the functional attributes and business category of the data processing task, the processing deadline indicates the time limit for the task to be completed within a specified period, and the resource requirements indicate the computing and network resources required to complete the task.

[0043] Specifically, the electronic device receives the target requirements submitted by the user, extracts the key information in the requirements through the task parsing module, and transforms it into specific data processing tasks. The task parsing module identifies the business type of data processing and determines the task type. It sets the processing deadline according to the time requirements in the requirements, and analyzes the complexity of the task and the data scale to estimate the required resource requirements, including computing resources and network resources.

[0044] For example, a distributed system receives a video analysis task, which is classified as an image processing task with a deadline of 30 minutes. Based on the amount of video data and the processing algorithm, it is estimated that the system requires 2 CPU cores and 4GB of memory, as well as 50Mbps of network bandwidth.

[0045] S103: Determine the task importance of each data processing task based on its task type, processing deadline, and resource requirements.

[0046] Specifically, after obtaining information on the type, deadline, and resource requirements of the data processing task, multiple indicators are comprehensively analyzed. If it is necessary to determine the importance of the task, it means that the current electronic device needs to evaluate the priority of the task. Therefore, the importance needs to be dynamically calculated, that is, the various characteristics of the task are weighted and evaluated. In this embodiment of the application, one method of calculating the importance is as follows: the system uses the following formula to calculate the task importance: Task importance = α1 × type weight + α2 × deadline weight + α3 × resource requirement, where α1, α2, and α3 are 0.4, 0.3, and 0.3 respectively. The type weight is divided according to the task type: 1.0 for real-time tasks, 0.8 for near-real-time tasks, and 0.6 for non-real-time tasks. The deadline weight is normalized (mapping the processing deadline to the 0-1 range, with shorter deadlines having higher weights). The resource requirement is a normalized value of resource usage such as CPU utilization and memory utilization. This allows for multi-dimensional task importance assessment during task scheduling. Reference Figure 6 , Figure 6 This is provided by the embodiments of this application. Figure 2 A flowchart illustrating a sub-step of step S103, including steps S501 to S504, is as follows: S501: Determine the basic sequence number of each data processing task based on the preset task type level table and the task type of each data processing task. The task type level table includes different basic sequence numbers corresponding to different task types.

[0047] Specifically, electronic devices need to determine the basic priority of tasks based on task type. This requires dynamically allocating basic sequence numbers, i.e., matching corresponding basic sequence numbers to data processing tasks. In this embodiment, one method of basic sequence number allocation is to pre-set a task type level table through a configuration interface. The corresponding relationships include: a basic sequence number of 1 for real-time tasks, a basic sequence number of 2 for ordinary tasks, and a basic sequence number of 3 for delayed tasks, thereby achieving an initial division of basic priorities during task processing. In other embodiments, the basic sequence number can also be dynamically adjusted based on the average execution time and resource consumption of different task types statistically analyzed from historical task execution data. The higher the execution overhead, the higher the corresponding basic sequence number. The basic sequence numbers in the task type level table can be set based on system operation and maintenance experience or obtained through performance testing. During testing, execution data of different task types is collected to analyze their resource consumption characteristics. When the system load is balanced, the mapping relationship between the corresponding task type and the basic sequence number is determined.

[0048] S502: Sort each data processing task in descending order of processing deadline, and determine the time sequence number of each data processing task. The earlier the task is sorted, the smaller its corresponding time sequence number.

[0049] Specifically, after the electronic device obtains the processing deadlines of each data processing task, it sorts the processing deadlines in descending order. If there are multiple tasks, it means that the current electronic device needs to determine the processing order based on the urgency of the tasks. Therefore, it is necessary to dynamically allocate time sequence numbers, that is, to match the corresponding time sequence number to the data processing tasks according to their sorting positions. In this embodiment, one method of time sequence number allocation is: the system directly uses the sorted position as the time sequence number, with the task with the longest processing deadline having a time sequence number of 1, the second longest task having a time sequence number of 2, and so on, thereby enabling priority division in the time dimension during task processing.

[0050] S503: Sort the data processing tasks according to their resource requirements from smallest to largest, and determine the resource sequence number of each data processing task. The earlier the task is sorted, the smaller its corresponding resource sequence number.

[0051] Specifically, after obtaining the resource requirements of each data processing task, the resource requirements are sorted in ascending order. If there are multiple tasks, it means that the current electronic device needs to determine the processing order based on resource consumption. Therefore, it is necessary to dynamically allocate resource numbers, that is, to match the corresponding resource numbers to the data processing tasks according to their sorting positions. In this embodiment, one resource number allocation method is: the system directly uses the sorted position as the resource number. The resource number of the task with the smallest resource requirement is 1, the resource number of the next smallest task is 2, and so on, thereby enabling priority division of resources during task processing.

[0052] S504 calculates the product of the base sequence number, time sequence number, and resource sequence number to obtain the task importance.

[0053] Specifically, after obtaining the basic sequence number, time sequence number, and resource sequence number of each data processing task, the three sequence numbers are multiplied together. If a product is obtained, it indicates that the current electronic device needs to determine the final priority based on multi-dimensional indicators. Therefore, the task importance needs to be dynamically calculated, that is, the sequence number product is calculated for the data processing tasks. In this embodiment, one method of calculating importance is: the system directly multiplies the basic sequence number, time sequence number, and resource sequence number to obtain the task importance. The larger the product, the higher the task importance, and the more priority it needs to be processed. This enables a multi-dimensional comprehensive priority evaluation during task processing.

[0054] S104: Calculate the processing suitability between each first edge device and each data processing task based on the importance of each task, the computing resource utilization rate of each first edge device, and the network bandwidth utilization rate.

[0055] Specifically, after obtaining the importance of each task and the resource status of the edge device, the various indicators of the task and the device are matched and analyzed. If it is necessary to calculate the processing adaptability, it means that the current electronic device needs to reasonably allocate tasks to the appropriate first edge device. Therefore, it is necessary to dynamically evaluate the matching degree between the task and the first edge device, that is, to calculate the processing adaptability. In this embodiment of the application, one way to calculate the processing adaptability is as follows: The system uses the following formula to calculate the processing adaptability score: Processing adaptability = α × task importance + β × (1 - computing resource utilization rate) + γ × (1 - network bandwidth utilization rate) where α, β, and γ are the importance weight, resource weight, and bandwidth weight, respectively, and α + β + γ = 1. The higher the score, the higher the matching degree between the task and the first edge device, thereby achieving optimal resource utilization in the task allocation process.

[0056] Reference Figure 3 , Figure 3 This is provided by the embodiments of this application. Figure 2 A flowchart illustrating a sub-step of step S104, including steps S201 to S204, is as follows: S201: Assign priority weights to each data processing task based on task importance.

[0057] Specifically, after obtaining the task importance of each data processing task, the importance value is normalized. If priority weights need to be assigned, it means that the current electronic device needs to allocate processing resources reasonably according to the importance of the tasks. Therefore, priority weights need to be dynamically calculated, that is, resource allocation ratios are set for data processing tasks. In this embodiment, one method of priority weight calculation is as follows: the system uses the following formula to calculate the weight value: priority weight = current task importance / sum of the importance of all tasks, thereby enabling differentiated allocation based on task importance during resource scheduling.

[0058] Reference Figure 4 , Figure 4 This is provided by the embodiments of this application. Figure 3 A flowchart illustrating a sub-step of step S201, including steps S301 to S304, is shown below: S301: Divide the importance of data processing tasks into multiple level intervals, with each interval corresponding to a different initial weight.

[0059] Specifically, after obtaining the task importance range of the data processing task, the task importance value is divided into different intervals. If it is necessary to determine the initial weight, it means that the current electronic device needs to be managed hierarchically according to the task importance. Therefore, it is necessary to dynamically match the weight value, that is, to set the corresponding initial weight for different level intervals. In this embodiment of the application, one interval division method is as follows: the system divides the task importance into three level intervals: high, medium and low. The task importance is greater than 0.8 and is a high-level task with an initial weight of 0.5. The task importance is between 0.4 and 0.8 and is a medium-level task with an initial weight of 0.3. The task importance is less than 0.4 and is a low-level task with an initial weight of 0.2. This enables the initial resource allocation based on the level during the task processing.

[0060] S302: Calculate the difference between the processing deadline of the data processing task and the current time to determine the time urgency.

[0061] Specifically, after obtaining the processing deadline of the data processing task and the current system time, the time difference between the two is calculated. If it is necessary to determine the time urgency, it means that the current electronic device needs to assess the urgency based on the remaining time of the task. Therefore, the urgency value needs to be dynamically calculated, that is, the remaining processing time of the task is calculated. In this embodiment of the application, one way to calculate the urgency is as follows: the system uses the following formula to calculate the time urgency: Time urgency = 1 - (processing deadline - current time) / preset maximum processing time, where the preset maximum processing time is 10 hours. The closer the urgency value is to 1, the more urgent the task is, thereby enabling priority assessment based on the time dimension during task processing.

[0062] S303: Determine the request priority based on the request frequency of the data processing task within a preset time period.

[0063] Specifically, after obtaining the number of requests for the data processing task within a preset time period, the number of requests is calculated in relation to the length of the time period. If it is necessary to determine the request priority, it means that the current electronic device needs to assess the importance of the task based on the access pattern. Therefore, it is necessary to dynamically calculate the request priority, that is, to count the request frequency per unit time. In this embodiment, one priority calculation method is as follows: the system uses the following formula to calculate the request priority: Request priority = Number of requests / Preset time period length, where the preset time period length is 1 hour. The higher the request priority, the more frequently the task is accessed, thereby enabling priority assessment based on access frequency during task processing.

[0064] S304: The initial weight, time urgency, and request priority are weighted to obtain the priority weight of the data processing task.

[0065] Specifically, after obtaining the initial weight, time urgency, and request priority of the data processing task, the three indicators are weighted and combined. If it is necessary to determine the final priority weight, it means that the current electronic device needs to evaluate the importance of the task based on multi-dimensional characteristics. Therefore, the priority weight needs to be dynamically calculated, that is, weight coefficients are set for the three indicators and weighted summation is performed. In this embodiment, one weight calculation method is as follows: the system uses the following formula to calculate the priority weight: priority weight = λ1 × initial weight + λ2 × time urgency + λ3 × request priority, where λ1, λ2, and λ3 are 0.3, 0.4, and 0.3, respectively. The larger the weight value, the higher the task processing priority, thereby enabling multi-dimensional comprehensive evaluation during task scheduling.

[0066] S202: Calculate the computing resource availability rate of the first edge device by calculating the computing resource occupancy rate. The computing resource availability rate is the ratio of unoccupied computing resources to the total computing resources.

[0067] Specifically, after obtaining the computing resource occupancy rate of the first edge device, the occupancy rate data is converted into availability rate. If it is necessary to determine the computing resource availability rate, it means that the current electronic device needs to evaluate the resource status of the first edge device. Therefore, it is necessary to dynamically calculate the availability rate, that is, calculate the proportion of unoccupied resources. In this embodiment, one method of availability rate calculation is as follows: The system uses the following formula to calculate the computing resource availability rate: Computing resource availability rate = (Total computing resources - Occupied computing resources) / Total computing resources = 1 - Computing resource occupancy rate, thereby enabling load assessment based on availability rate during resource scheduling.

[0068] S203: Calculate the network transmission availability of the first edge device by the network bandwidth utilization rate. The network transmission availability rate is the ratio of unused bandwidth resources to the total bandwidth resources.

[0069] Specifically, after obtaining the network bandwidth occupancy rate of the first edge device, the occupancy rate data is converted into availability. If it is necessary to determine the network transmission availability rate, it means that the current electronic device needs to evaluate the bandwidth status of the first edge device. Therefore, it is necessary to dynamically calculate the availability rate, that is, calculate the proportion of unoccupied bandwidth. In this embodiment, one method of availability rate calculation is as follows: the system uses the following formula to calculate the network transmission availability rate: network transmission availability rate = (total bandwidth resources - occupied bandwidth resources) / total bandwidth resources = 1 - network bandwidth occupancy rate, thereby enabling network evaluation based on availability rate during resource scheduling.

[0070] S204: Based on priority weights and the availability of computing resources and network transmission of edge devices, determine the processing suitability of each first edge device and each data processing task.

[0071] Specifically, after obtaining the priority weight of the data processing task and the resource status of the edge device, the adaptability of each indicator is calculated. If it is necessary to determine the processing adaptability, it means that the current electronic device needs to evaluate the matching degree between the task and the device. Therefore, the adaptability needs to be dynamically calculated, that is, the task priority and the status of the first edge device are considered comprehensively. In this embodiment of the application, one adaptability calculation method is as follows: The system uses the following formula to calculate the processing adaptability: Processing adaptability = μ1 × priority weight + μ2 × computing resource availability + μ3 × network transmission availability, where μ1, μ2, and μ3 are 0.4, 0.3, and 0.3, respectively. The higher the adaptability, the more suitable the task is to be processed on the corresponding first edge device, thereby achieving optimal resource matching in the task allocation process.

[0072] Reference Figure 5 , Figure 5 This is provided by the embodiments of this application. Figure 3 A flowchart illustrating a sub-step of step S204, including steps S401 to S404, is as follows: S401: Calculate the ratio of the computing resource availability of each first edge device to the resource requirements of each data processing task to obtain the computing resource matching degree.

[0073] Specifically, after obtaining the computing resource availability rate of the first edge device and the resource requirements of the data processing task, the ratio of availability rate to requirements is calculated. If it is necessary to determine the computing resource matching degree, it means that the current electronic device needs to evaluate the resource supply and demand relationship, so the matching degree needs to be dynamically calculated, that is, the computing resource supply and demand ratio. In this embodiment of the application, one way to calculate the matching degree is as follows: the system uses the following formula to calculate the computing resource matching degree: computing resource matching degree = computing resource availability rate / task resource requirement. A matching degree greater than 1 indicates that the available resources of the first edge device meet the task requirements, thereby enabling resource-based adaptability assessment during task allocation.

[0074] S402: Calculate the ratio of the network transmission availability of each first edge device to the data transmission requirements of the data processing task to obtain the network resource matching degree.

[0075] Specifically, after obtaining the network transmission availability rate and data processing task transmission requirements of the first edge device, the ratio of availability rate to requirements is calculated. If it is necessary to determine the network resource matching degree, it means that the current electronic device needs to evaluate the network supply and demand relationship. Therefore, the matching degree needs to be dynamically calculated, that is, the network supply and demand ratio is calculated. In this embodiment, one way to calculate the matching degree is as follows: the system uses the following formula to calculate the network resource matching degree: network resource matching degree = network transmission availability rate / task transmission requirement. A matching degree greater than 1 indicates that the available bandwidth of the first edge device meets the task requirements, thereby enabling network-based adaptability assessment during task allocation.

[0076] S403: Based on the preset weights for matching computational resources and network resources, the weighted sum of the matching degree of computational resources and network resources is obtained to get the total resource matching degree.

[0077] Specifically, after obtaining the computing resource matching degree and the network resource matching degree, the two matching degrees are weighted and combined. If it is necessary to determine the total resource matching degree, it means that the current electronic device needs to evaluate the overall resource adaptability. Therefore, the total matching degree needs to be dynamically calculated, that is, the matching degree of different dimensions is weighted and summed. In this embodiment of the application, one way to calculate the total matching degree is as follows: the system uses the following formula to calculate the total resource matching degree: total resource matching degree = ω1 × computing resource matching degree + ω2 × network resource matching degree, where ω1 and ω2 are 0.6 and 0.4, respectively. The higher the total matching degree, the more suitable the overall resources of the first edge device are for processing the task, thereby enabling multi-dimensional resource evaluation in the task allocation process.

[0078] S404: Calculate the product of the total resource matching degree and the task priority weight to obtain the processing adaptability between each first edge device and each data processing task.

[0079] Specifically, after obtaining the total resource matching degree and task priority weight, the two indicators are multiplied. If it is necessary to determine the processing adaptability, it means that the current electronic device needs to comprehensively evaluate the matching relationship between the task and the first edge device. Therefore, the processing adaptability needs to be dynamically calculated, that is, the resource adaptability is combined with the importance of the task. In this embodiment, one way to calculate the adaptability is as follows: The system uses the following formula to calculate the processing adaptability: Processing adaptability = Total resource matching degree × Task priority weight. The higher the processing adaptability, the better the processing effect of the task on the corresponding first edge device, thereby enabling a comprehensive evaluation of resources and priorities in the task allocation process.

[0080] S105: Determine the stability of each first edge device based on its historical failure rate and runtime.

[0081] Specifically, after acquiring the historical fault data and cumulative running time of the first edge device, the fault situation and running time are comprehensively analyzed. If it is necessary to determine the stability of the first edge device, it means that the reliability of the first edge device needs to be evaluated by the current electronic device. Therefore, it is necessary to dynamically calculate the stability index, that is, calculate the fault probability per unit time. In this embodiment of the application, one stability calculation method is as follows: the system uses the following formula to calculate the device stability: device stability = (1 - number of historical faults / running time) × e^(-running time / preset lifespan), where the preset lifespan is 10,000 hours. The closer the stability is to 1, the more reliable the first edge device is, thereby enabling the device evaluation based on reliability in the task allocation process.

[0082] Reference Figure 7 , Figure 7 This is provided by the embodiments of this application. Figure 2 A flowchart illustrating a sub-step of step S105, including steps S601 to S603, is as follows: S601: Calculate the ratio of the historical failure rate to the historical failure rate threshold, and use the reciprocal of the ratio as the first stability value.

[0083] Specifically, after obtaining the historical failure rate and the preset failure rate threshold of the first edge device, the ratio of the historical failure rate to the threshold is calculated. If it is necessary to determine the first stability value, it means that the current electronic device needs to evaluate the failure status of the first edge device. Therefore, it is necessary to dynamically calculate the stability, that is, calculate the relative level of the failure rate. In this embodiment, one stability calculation method is as follows: the system uses the following formula to calculate the first stability value: first stability value = historical failure rate threshold / historical failure rate, where the historical failure rate threshold is 0.1. The larger the first stability value, the lower the failure rate of the first edge device, thereby enabling a reliability measurement based on the failure rate to be realized in the evaluation process of the first edge device.

[0084] S602: Calculate the ratio of runtime to runtime threshold, and use the ratio as the second stability value.

[0085] Specifically, after obtaining the runtime of the first edge device and the preset runtime threshold, the runtime is compared with the threshold. If a second stability value needs to be determined, it means that the current electronic device needs to evaluate the device's operating status. Therefore, stability needs to be dynamically calculated, that is, the relative level of runtime is calculated. In this embodiment, one stability calculation method is as follows: the system uses the following formula to calculate the second stability value: second stability value = runtime / runtime threshold, where the runtime threshold is 1000 hours. The larger the second stability value, the richer the operating experience of the first edge device, thereby enabling a reliability measurement based on runtime in the evaluation process of the first edge device.

[0086] S603: The first stability value and the second stability value are respectively weighted and summed with the corresponding preset weights to obtain the stability of each first edge device.

[0087] Specifically, after obtaining the first stability value and the second stability value, the two stability indicators are weighted and combined. If it is necessary to determine the stability of the first edge device, it means that the current electronic device needs to evaluate the overall reliability of the first edge device. Therefore, it is necessary to dynamically calculate the stability, that is, to sum the stability of different dimensions. In this embodiment, one way to calculate the stability is as follows: the system uses the following formula to calculate the device stability: device stability = θ1 × first stability value + θ2 × second stability value, where θ1 and θ2 are 0.6 and 0.4, respectively. The higher the stability, the better the overall reliability of the first edge device, thereby enabling multi-dimensional reliability assessment during task allocation.

[0088] S106: Based on the stability and processing adaptability of each first edge device, determine the second edge device, and process the data processing task through the second edge device.

[0089] Specifically, after obtaining the stability and processing adaptability of each first edge device, the two indicators are comprehensively evaluated. If it is necessary to determine the second edge device, the second edge device is the first edge device with the best comprehensive evaluation of stability and processing adaptability among the first edge devices. Then, the device needs to be selected dynamically. In this embodiment, one device selection method is as follows: the system calculates the device comprehensive score using the following formula: comprehensive score = ρ1 × stability + ρ2 × processing adaptability, where ρ1 and ρ2 are 0.4 and 0.6 respectively. The first edge device with the highest score is selected as the second edge device to perform the data processing task, thereby enabling reliable and efficient resource scheduling during task execution.

[0090] Reference Figure 8 , Figure 8 This is provided by the embodiments of this application. Figure 2A flowchart illustrating a sub-step of step S106, including steps S701 to S703, is as follows: S701: Select the first edge device with stability greater than the stability threshold and processing fit greater than the processing fit threshold to obtain candidate edge devices.

[0091] Specifically, after obtaining the stability and processing adaptability of the first edge device, the various indicators are compared with preset thresholds. If it is necessary to determine candidate edge devices, it means that the current electronic device needs to be preliminarily screened for qualified devices. Therefore, the first edge device needs to be dynamically screened, that is, a dual threshold constraint condition is set. In this embodiment of the application, one screening method is: the system sets the stability threshold to 0.8 and the processing adaptability threshold to 0.7. The first edge device that meets both conditions is used as a candidate edge device.

[0092] S702: The stability and processing adaptability are weighted and summed with their respective preset weights to obtain the comprehensive score of each candidate edge device.

[0093] Specifically, after obtaining the stability and processing adaptability of candidate edge devices, the two indicators are weighted and combined. If a comprehensive score needs to be determined, it means that the current electronic device needs to evaluate the overall performance of the device. Therefore, the score needs to be dynamically calculated, that is, the weighted sum of various indicators of the candidate edge devices. In this embodiment, one way to calculate the score is as follows: The system uses the following formula to calculate the comprehensive score: Comprehensive score = δ1 × stability + δ2 × processing adaptability, where δ1 and δ2 are 0.4 and 0.6, respectively. The higher the score, the more suitable the candidate edge device is for performing the task.

[0094] S703: The candidate edge device with the highest overall score is identified as the second edge device.

[0095] Specifically, after obtaining the comprehensive scores of the candidate edge devices, the scores are compared. If it is necessary to determine the second edge device, it means that the current electronic device needs to select the optimal execution device. Therefore, the device needs to be selected dynamically, that is, the candidate edge device with the highest score is selected from the candidate devices. In this embodiment, one selection method is: the system directly selects the candidate edge device with the highest comprehensive score as the second edge device, thereby achieving the optimal device selection in the task allocation process.

[0096] Figure 9 This is a schematic diagram of a distributed data processing system for edge intelligence disclosed in an embodiment of this application, including: Data acquisition module 21 is used to acquire status data of at least one first edge device, the status data including computing resource utilization rate, network bandwidth utilization rate, historical failure rate and runtime of the first edge device; The task determination module 22 is used to determine at least one data processing task based on the target requirements. The data processing task includes the task type, processing period and resource requirements of the data processing task. The importance assessment module 23 is used to determine the importance of the corresponding data processing task based on the task type, processing period, and resource requirements of each data processing task. The adaptation calculation module 24 is used to calculate the processing adaptation between each first edge device and each data processing task based on the importance of each task, the computing resource utilization rate of each first edge device, and the network bandwidth utilization rate. Stability assessment module 25 is used to determine the stability of each of the first edge devices based on the historical failure rate and the runtime of the first edge devices; The device selection module 26 is used to determine a second edge device based on the stability of each of the first edge devices and the processing adaptability, and to process the data processing task through the second edge device.

[0097] The task determination module 22 is further configured to: determine the basic sequence number of each data processing task according to a preset task type level table and the task type of each data processing task; the task type level table includes different basic sequence numbers corresponding to different task types; sort each data processing task from largest to smallest processing period to determine the time sequence number of each data processing task, with the earlier the sorting, the smaller the corresponding time sequence number; sort the data processing tasks from smallest to largest resource requirement to determine the resource sequence number of each data processing task, with the earlier the sorting, the smaller the corresponding resource sequence number; and calculate the product of the basic sequence number, the time sequence number, and the resource sequence number to obtain the task importance.

[0098] The adaptation calculation module 24 is further configured to: assign priority weights to each data processing task based on the task importance; calculate the computing resource availability rate of the first edge device through the computing resource occupancy rate, wherein the computing resource availability rate is the ratio of unoccupied computing resources to the total computing resources; calculate the network transmission availability rate of the first edge device through the network bandwidth occupancy rate, wherein the network transmission availability rate is the ratio of unoccupied bandwidth resources to the total bandwidth resources; and determine the processing adaptation rate between each first edge device and each data processing task based on the priority weights and the computing resource availability rate and network transmission availability rate of the edge device.

[0099] The adaptation calculation module 24 is further configured to calculate the ratio of the computing resource availability of each first edge device to the resource requirement of each data processing task to obtain the computing resource matching degree; calculate the ratio of the network transmission availability of each first edge device to the data transmission requirement of the data processing task to obtain the network resource matching degree; according to the preset computing resource matching degree weight and network resource matching degree weight, weighted summation of the computing resource matching degree and the network resource matching degree to obtain the total resource matching degree; and calculate the product of the total resource matching degree and the task priority weight to obtain the processing adaptation degree between each first edge device and each data processing task.

[0100] The adaptability calculation module 24 is also used to divide the importance of the data processing task into multiple level intervals, each interval corresponding to a different initial weight; calculate the difference between the processing deadline of the data processing task and the current time to determine the time urgency; determine the request priority based on the request frequency of the data processing task within a preset time period; and perform a weighted calculation on the initial weight, the time urgency, and the request priority to obtain the priority weight of the data processing task.

[0101] The stability assessment module 25 is further configured to calculate the ratio of the historical failure rate to the historical failure rate threshold, and use the reciprocal of the ratio as the first stability value; calculate the ratio of the runtime to the runtime threshold, and use the ratio as the second stability value; and sum the first stability value and the second stability value with their respective preset weights to obtain the stability of each of the first edge devices.

[0102] The device selection module 26 is further configured to filter first edge devices whose stability is greater than the stability threshold and whose processing adaptability is greater than the processing adaptability threshold, to obtain candidate edge devices; to perform weighted summation of the stability and the processing adaptability with their respective preset weights, to obtain a comprehensive score for each candidate edge device; and to determine the candidate edge device with the highest comprehensive score as the second edge device.

[0103] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0104] This embodiment also discloses an electronic device 900, as shown in the reference... Figure 10The electronic device may include: at least one processor 901, at least one communication bus 902, user interface 903, network interface 904, and at least one memory 905.

[0105] The communication bus 902 is used to enable communication between these components.

[0106] The user interface 903 may include a display screen and a camera. Optional user interfaces may also include standard wired interfaces and wireless interfaces.

[0107] The network interface 904 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0108] The processor 901 may include one or more processing cores. The processor connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory, and by calling data stored in memory. Optionally, the processor may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor.

[0109] The memory 905 may include random access memory (RAM) or read-only memory. Optionally, the memory may include a non-transitory computer-readable storage medium. The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor. As shown in the figure, the memory, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for an edge intelligence distributed data processing method.

[0110] exist Figure 10 In the electronic device shown, the user interface is mainly used to provide an input interface for the user and obtain the user input data; while the processor can be used to call an application program that stores an edge intelligence distributed data processing method in the memory. When executed by one or more processors, the electronic device executes one or more methods as described in the above embodiments.

[0111] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0112] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0114] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0115] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0116] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0117] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the disclosure in this specification. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method of edge intelligent distributed data processing, characterized in that, The method comprises: obtaining state data of at least one first edge device, the state data comprising a computing resource occupation rate, a network bandwidth occupation rate, a historical failure rate and a running time length of the first edge device; determining at least one data processing task according to a target requirement, the data processing task comprising a task type, a processing deadline and a resource requirement of the data processing task; determining a task importance degree of a corresponding data processing task based on the task type, the processing deadline and the resource requirement of each data processing task; calculating a processing adaptation degree of each first edge device and each data processing task respectively based on the task importance degree, the computing resource occupation rate and the network bandwidth occupation rate of each first edge device; determining a stability of each first edge device based on the historical failure rate and the running time length of the first edge device; determining a second edge device based on the stability and the processing adaptation degree of each first edge device, and performing data processing task processing through the second edge device.

2. The method of claim 1, wherein, The task importance degree, the computing resource occupation rate and the network bandwidth occupation rate of each first edge device are determined, and the processing adaptation degree of each first edge device and each data processing task is determined, specifically comprising: allocating a priority weight for each data processing task based on the task importance degree; calculating a computing resource availability rate of the first edge device through the computing resource occupation rate, the computing resource availability rate being a ratio of unoccupied computing resources to total computing resources; calculating a network transmission availability rate of the first edge device through the network bandwidth occupation rate, the network transmission availability rate being a ratio of unoccupied bandwidth resources to total bandwidth resources; determining the processing adaptation degree of each first edge device and each data processing task based on the priority weight and the computing resource availability rate and the network transmission availability rate of the edge device.

3. The method of claim 2, wherein, The priority weight and the computing resource availability rate and the network transmission availability rate of the first edge device are determined, and the processing adaptation degree of each first edge device and each data processing task is determined, specifically comprising: calculating a ratio of the computing resource availability rate of each first edge device to the resource requirement of each data processing task to obtain a computing resource matching degree; calculating a ratio of the network transmission availability rate of each first edge device to the data transmission requirement of the data processing task to obtain a network resource matching degree; weighting and summing the computing resource matching degree and the network resource matching degree according to preset computing resource matching degree weights and network resource matching degree weights to obtain a resource total matching degree; calculating a product of the resource total matching degree and the task priority weight to obtain the processing adaptation degree of each first edge device and each data processing task.

4. The method of claim 2, wherein, The priority weight for each data processing task is allocated based on the task importance degree, specifically comprising: dividing the data processing task importance degree into multiple level intervals, each interval corresponding to a different initial weight; calculating a difference between the processing deadline of the data processing task and the current time to determine a time urgency degree; determine a request priority according to a request frequency of the data processing task within a preset time period; perform weighted calculation on the initial weight, the time urgency and the request priority to obtain a priority weight of the data processing task.

5. The method of claim 1, wherein, The task importance of each data processing task is determined based on the data processing task type, processing deadline and resource requirement, specifically including: determine a basic serial number of each data processing task according to a preset task type level table and the task type of each data processing task, wherein the task type level table includes different basic serial numbers corresponding to different task types; sort each data processing task according to the processing deadline from large to small to determine a time serial number of each data processing task, and the earlier the sorting, the smaller the corresponding time serial number; sort each data processing task according to the resource requirement from small to large to determine a resource serial number of each data processing task, and the earlier the sorting, the smaller the corresponding resource serial number; calculate the product of the basic serial number, the time serial number and the resource serial number to obtain the task importance.

6. The method of claim 1, wherein, The stability of each first edge device is determined based on the historical failure rate and the running length of the first device, specifically including: calculate the ratio of the historical failure rate to a historical failure rate threshold value, and take the inverse of the ratio as the first stability value; calculate the ratio of the running length to a running length threshold value, and take the ratio as the second stability value; weight and sum the first stability value and the second stability value respectively with corresponding preset weights to obtain the stability of each first edge device.

7. The method of claim 1, wherein, The second edge device is determined based on the stability of each first edge device and the processing adaptation degree, specifically including: screen the first edge device with the stability greater than the stability threshold value and the processing adaptation degree greater than the processing adaptation degree threshold value to obtain a candidate edge device; weight and sum the stability and the processing adaptation degree respectively with corresponding preset weights to obtain a comprehensive score of each candidate edge device; determine the candidate edge device with the highest comprehensive score as the second edge device.

8. An edge intelligent distributed data processing system, characterized by, including: a data acquisition module configured to acquire state data of at least one first edge device, the state data including a computing resource occupation rate, a network bandwidth occupation rate, a historical failure rate and a running length of the first edge device; a task determination module configured to determine at least one data processing task according to a target requirement, the data processing task including a task type, a processing deadline and a resource requirement of the data processing task; an importance evaluation module configured to determine a task importance of a corresponding data processing task based on the task type, the processing deadline and the resource requirement of each data processing task; an adaptation degree calculation module configured to calculate a processing adaptation degree of each first edge device and each data processing task according to each task importance, the computing resource occupation rate and the network bandwidth occupation rate of each first edge device; a stability evaluation module configured to determine a stability of each of the first edge devices based on the historical failure rate and the runtime length of the first edge devices; a device selection module configured to determine a second edge device based on the stability of each of the first edge devices and the processing fitness, and to perform the data processing task by the second edge device.

9. An electronic device, comprising: An electronic device including a processor, a memory, a user interface, and a network interface, the memory configured to store instructions, the user interface and the network interface each configured to communicate with other devices, and the processor configured to execute the instructions stored in the memory to cause the electronic device to perform the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, A computer-readable storage medium storing instructions that, when executed, perform the method of any one of claims 1-7.

Images