A collaborative processing method and device based on distributed edge computing and a medium

Abstract

The application discloses a kind of based on distributed edge computing's collaborative processing method, device and medium, first loss value of first data processing model based on distributed edge cooperation is calculated;Determine maximum loss value and minimum loss value;If it is determined that first loss value is less than or equal to minimum loss value, keep the connection of terminal and edge server;If it is determined that first loss value is greater than minimum loss value, and less than maximum loss value, disconnect the connection of terminal and edge server, connect the edge server with the highest score with the terminal;If it is determined that first loss value is greater than or equal to maximum loss value, the edge server with the highest comprehensive satisfaction is added to the connection with the terminal.The application proposes a kind of based on distributed edge computing's collaborative processing method, device and medium, by the better edge server of performance selection and reduce the difference between edge servers Service processing performance, improve the ability of multiple edge servers collaborative processing.

Description

Technical Field

[0001] This invention relates to the field of edge computing communication technology, and in particular to a collaborative processing method, apparatus and medium based on distributed edge computing. Background Technology

[0002] With the gradual integration of various new energy-consuming devices such as distributed energy resources, energy storage, and loads, and the increasing demands for real-time monitoring of energy consumption during digital transformation, a large number of data acquisition terminals need to be deployed to collect information from distribution areas at high frequency. This large-scale deployment of acquisition terminals leads to a surge in data volume, and consequently increases the requirements for collaborative processing capabilities between edge servers connected to different terminals.

[0003] In existing distributed edge computing collaboration methods, the overall business processing performance of the edge server collaboration set is poor, and it is impossible to select the best edge server based on the performance of each edge server. This results in large differences in business processing performance between servers, low efficiency of collaborative processing among multiple edge servers, and low ability of edge servers to collaboratively process high-frequency acquisition terminal data. Summary of the Invention

[0004] This invention provides a collaborative processing method, apparatus, and medium based on distributed edge computing to solve the problem that existing technologies cannot effectively improve the collaborative processing capabilities of multiple edge servers.

[0005] To address the above problems, this invention provides a collaborative processing method based on distributed edge computing, comprising:

[0006] Calculate the first loss value of the power service processing performance of the distribution area based on the first data processing model of distributed edge collaboration, wherein the data processing model is a collaborative set composed of several edge servers connected to the terminal.

[0007] Determine the power service processing performance loss evaluation threshold for any edge server in the first data processing model to the first terminal connected to it, wherein the upper and lower limits of the threshold are the maximum loss value and the minimum loss value, respectively.

[0008] If it is determined that the first loss value is less than or equal to the minimum loss value, maintain the connection between the first terminal and the first edge server;

[0009] If it is determined that the first loss value is greater than the minimum loss value and less than the maximum loss value, disconnect the first terminal from the first edge server, calculate the score of each edge server in the preset edge server set, and connect the second edge server with the highest score to the first terminal.

[0010] If it is determined that the first loss value is greater than or equal to the maximum loss value, the connection between the first terminal and the first edge server is maintained, the overall satisfaction of each edge server in the preset edge server set is calculated, and the third edge server with the highest overall satisfaction is added to the connection with the first terminal.

[0011] This invention provides an objective standard for evaluating the performance of edge servers by determining a threshold for evaluating the performance loss of power service processing at terminals. By comparing the first loss value with the maximum and minimum loss values, different methods can be used to select new edge servers or retain existing ones under different circumstances. Servers with poor service processing performance can be eliminated, while servers with good service processing performance can be retained. Based on the comprehensive satisfaction among edge servers, the service data processing models of multiple edge servers can be aggregated to improve the ability of multiple edge servers to collaboratively process high-frequency acquisition terminal data. By controlling the average collaboration index to be greater than or equal to the average collaboration threshold and the standard deviation collaboration index to be greater than or equal to the standard deviation collaboration threshold, the overall service processing performance of the edge server collaboration set can be improved, and the differences in service processing performance between edge servers can be reduced.

[0012] As a preferred approach, the first loss value for the power service processing performance of the distribution area based on the distributed edge collaborative data processing model is calculated as follows:

[0013] Based on the data transmission latency, computing resources, and relative load balancing of the distribution terminal, the performance evaluation indicators for power service processing are determined.

[0014] The expected performance index of power business processing is calculated by power business processing performance evaluation index. Based on the power business processing performance evaluation index and the expected performance index of power business processing, the first loss value of power business processing performance of the distribution area is obtained by weighting.

[0015] This preferred scheme calculates and determines the performance evaluation index for power service processing. The larger the value, the lower the transmission latency and computation latency of the edge server, indicating better edge server performance. The expected performance index for power service processing is obtained through the power service processing performance evaluation index. Then, the first loss value is calculated using the power service processing performance evaluation index and the expected performance index for power service processing. The first loss value obtained can be used as a standard for whether the edge server can tolerate the performance loss of power service processing of the terminal, providing an objective measure for whether the edge server meets the conditions.

[0016] As a preferred approach, the score of each edge server in the preset edge server set is calculated, specifically as follows:

[0017] Based on the bandwidth and channel gain between the first terminal and each edge server in the preset edge server set, the first terminal calculates its score for each edge server in the preset edge server set, thus obtaining a first score. .

[0018] This preferred solution calculates a first score. The higher the first score of the terminal for the edge server, the better the data processing performance of the edge server. The terminal selects the edge server with the highest first score for data processing. The edge server with the highest first score can optimize the local business data processing model and improve edge processing performance.

[0019] As a preferred embodiment, the overall satisfaction level of each edge server in the preset edge server set is calculated, specifically as follows:

[0020] Based on the transmission distance, energy consumption, and model transmission rate between different edge servers in the preset edge server set, the overall satisfaction of each edge server in the preset edge server set is calculated to obtain the first satisfaction level.

[0021] The first satisfaction level is:

[0022]

[0023] in, and These are preset edge servers and centralized edge servers. and The transmission distance and energy consumption of the business data processing model between them. To pre-configure edge servers and centralize edge servers and Between the models, transmission rate To pre-configure edge servers and centralize edge servers Energy consumed in business data processing Pre-set edge servers and centralized edge servers before transmitting business data processing models. Remaining energy.

[0024] This preferred solution calculates the first satisfaction level. The higher the first satisfaction level, the shorter the model transmission time between servers, and the lower the energy consumption for server model transmission and computation, indicating that the data processing capability of the edge server will be stronger. It can aggregate the business data processing models of multiple edge servers based on the first satisfaction level between edge servers, thereby improving the ability of multiple edge servers to collaboratively process high-frequency acquisition terminal data.

[0025] This invention provides a coordination capability verification method based on distributed edge computing, comprising:

[0026] Based on the collaborative processing method of distributed edge computing as described above, the terminals in the first data processing model are coordinated and processed, and the connection relationship between each edge server and each terminal in the first data processing model is adjusted to obtain the second data processing model.

[0027] Calculate the average synergy index and standard deviation synergy index of the second data processing model, and determine the average synergy threshold and standard deviation synergy threshold. Compare the average synergy index with the average synergy threshold and the standard deviation synergy index with the standard deviation synergy threshold.

[0028] The second data processing model is validated based on the comparison results until the average synergy index is greater than or equal to the average synergy threshold and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold.

[0029] This method, by controlling the average collaboration index to be greater than or equal to the average collaboration threshold and the standard deviation collaboration index to be greater than or equal to the standard deviation collaboration threshold, can eliminate servers with poor business processing performance and retain servers with good business processing performance, thereby improving the overall business processing performance of the edge server collaboration set and reducing the differences in business processing performance among edge servers.

[0030] As a preferred approach, the second data processing model is validated based on the comparison results until the average synergy index is greater than or equal to the average synergy threshold, and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold, specifically:

[0031] If the average synergy index is determined to be greater than or equal to the average synergy threshold, and the standard deviation synergy index is determined to be greater than or equal to the standard deviation synergy threshold, then the calculation ends.

[0032] If it is determined that the average coordination index is less than the average coordination threshold, or the standard deviation coordination index is less than the standard deviation coordination threshold, then the scores of each edge server in the second data processing model are recalculated. Based on the scoring results, the connection between the fourth edge server with the lowest score in the second data processing model and the terminal connected to it is disconnected. The second data processing model is updated based on the comparison results until the average coordination index is greater than or equal to the average coordination threshold and the standard deviation coordination index is greater than or equal to the standard deviation coordination threshold.

[0033] As a preferred approach, the average synergy index of the second data processing model is calculated as follows:

[0034] Based on the transmission latency and calculation latency of the business data processing model between each edge server in the first collaborative set of the second data processing model, the average collaboration degree index of the collaborative set composed of several edge servers connected to several terminals is obtained.

[0035] The average degree of synergy index is:

[0036]

[0037] in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

[0038] This preferred solution calculates the average collaboration index. The higher the average collaboration index, the lower the average business data processing latency of the edge servers within the collaboration set, and the better the overall business processing performance of the edge servers within the collaboration set.

[0039] As a preferred approach, the standard deviation coherence index of the second data processing model is calculated as follows:

[0040] Based on the transmission latency and computation latency of the business data processing model between each edge server in the first collaborative set of the second data processing model, calculate the standard deviation of the collaborative set formed by the several edge servers connected to several terminals.

[0041] The standard deviation synergy index is:

[0042]

[0043] in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

[0044] This preferred solution calculates the standard deviation synergy index, which is the reciprocal of the standard deviation of the business data processing latency of the edge server synergy set. It indicates the degree of fluctuation in the processing latency of the edge servers within the synergy set. The smaller the fluctuation, the smaller the performance difference of the edge servers within the synergy set.

[0045] This invention provides a collaborative processing device based on distributed edge computing, comprising:

[0046] The first loss value calculation module is used to calculate the first loss value of the power business processing performance of the distribution area based on the first data processing model of distributed edge collaboration, wherein the data processing model is a collaborative set composed of several edge servers connected to the terminal.

[0047] The second loss value determination module is used to determine the power service processing performance loss evaluation threshold of any edge server in the first data processing model for the first terminal connected to it. The upper and lower limits of the threshold are the maximum loss value and the minimum loss value, respectively. The first loss value is compared with the maximum loss value and the minimum loss value.

[0048] The first determination module is used to maintain the connection between the first terminal and the first edge server if it is determined that the first loss value is less than or equal to the minimum loss value.

[0049] The second judgment module is used to disconnect the connection between the first terminal and the second edge server if it is determined that the first loss value is greater than the minimum loss value and less than the maximum loss value, calculate the score of each edge server in the preset edge server set, and connect the third edge server with the highest score to the first terminal.

[0050] The third judgment module is used to calculate the overall satisfaction of each edge server in the preset edge server set if it is determined that the first loss value is greater than or equal to the maximum loss value, and add the fourth edge server with the highest overall satisfaction to the connection with the first terminal.

[0051] The present invention provides a storage medium storing a computer program, which is invoked and executed by a computer to implement the collaborative processing method based on distributed edge computing or the coordination capability verification method based on distributed edge computing as described above. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating a collaborative processing method based on distributed edge computing provided in an embodiment of the present invention.

[0053] Figure 2 This is a flowchart illustrating a coordination capability verification method based on distributed edge computing provided in an embodiment of the present invention.

[0054] Figure 3 This is a schematic diagram of a collaborative processing device based on distributed edge computing provided in an embodiment of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of this application 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. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0056] In the description of this application, it should be understood that the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," "third," or "fourth" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0057] Please see Figure 1 The present invention provides a collaborative processing method based on distributed edge computing, comprising:

[0058] S1. Calculate the first loss value of the power business processing performance of the distribution area based on the first data processing model of distributed edge collaboration, wherein the data processing model is a collaborative set composed of several edge servers connected to the terminal.

[0059] In this embodiment, a set of high-frequency acquisition terminals is defined: The edge server collaboration set connected to the terminal is the second collaboration set: Preset edge server set: ;

[0060] Based on the data transmission latency, computing resources, and relative load balancing of the distribution terminal, the performance evaluation indicators for power service processing are determined.

[0061] The performance evaluation indicators for power business processing are:

[0062]

[0063] Where e is the natural constant (approximately 2.718). For delay weighting, For data acquisition terminal Upload the business to the second collaborative centralized edge server Transmission delay, For the second collaborative centralized edge server Processing data from the data acquisition terminal The computational latency of the service As edge load weight, For the terminal The amount of cached business data, For the second collaborative centralized edge server The amount of cached business data, This refers to the number of high-frequency data acquisition terminals. This represents the number of edge servers in the second collaboration set.

[0064] The expected performance index of power business processing is obtained by averaging the performance evaluation indexes of power business processing. ;

[0065] Based on the power business processing performance evaluation index and the expected power business processing performance index, the first loss value of the power business processing performance of the distribution area is obtained by weighting.

[0066] The first loss value is:

[0067]

[0068] in, The weights of the actual loss evaluation indicators, As an evaluation indicator for power business processing performance, Expected performance indicators for power business processing.

[0069] This embodiment calculates and determines the performance evaluation index for power service processing. The larger the value, the lower the transmission latency and computation latency of the edge server, indicating better performance of the edge server. The expected performance index for power service processing is obtained through the power service processing performance evaluation index. Then, the first loss value is calculated using the power service processing performance evaluation index and the expected performance index for power service processing. The first loss value obtained can be used as a standard for whether the edge server can tolerate the power service processing performance loss of the terminal, providing an objective measurement scale for whether the edge server meets the conditions.

[0070] S2. Determine the power service processing performance loss evaluation threshold for any edge server in the first data processing model to the first terminal connected to it. The upper and lower limits of the threshold are the maximum loss value and the minimum loss value, respectively.

[0071] In this embodiment, the maximum loss value and the minimum loss value are determined as follows: and .

[0072] S3. If it is determined that the first loss value is less than or equal to the minimum loss value, maintain the connection between the first terminal and the first edge server;

[0073] In this embodiment, if it is determined that the first loss value is less than the minimum loss value ( Maintain the connection between the first terminal and the first edge server;

[0074] It should be noted that the situation in this embodiment refers to the first edge server. For the terminal If the performance loss in power service processing is tolerable, then the first edge server will continue to be selected. Perform data processing.

[0075] S4. If it is determined that the first loss value is greater than the minimum loss value and less than the maximum loss value, disconnect the connection between the first terminal and the first edge server, calculate the score of each edge server in the preset edge server set, and connect the second edge server with the highest score to the first terminal.

[0076] In this embodiment, if it is determined that the first loss value is greater than the minimum loss value and less than the maximum loss value ( ), disconnect the first terminal from the first edge server;

[0077] Based on the bandwidth and channel gain between the first terminal and each edge server in the preset edge server set, the score of the first terminal on each edge server in the preset edge server set is calculated, and the first score is obtained. ;

[0078] Connect the second edge server with the highest first score to the first terminal.

[0079] It should be noted that the scenario in this embodiment represents a significant performance loss for the terminal. The performance loss needs to be reduced by selecting a different server.

[0080] In this embodiment, a first score is calculated. The higher the first score of the terminal for the edge server, the better the data processing performance of the edge server. The terminal selects the edge server with the highest first score for data processing. The edge server with the highest first score can optimize the local business data processing model and improve edge processing performance.

[0081] S5. If the first loss value is determined to be greater than or equal to the maximum loss value, maintain the connection between the first terminal and the first edge server, calculate the overall satisfaction of each edge server in the preset edge server set, and add the third edge server with the highest overall satisfaction to the connection with the first terminal.

[0082] In this embodiment, if it is determined that the first loss value is greater than or equal to the maximum loss value ( Maintain the connection between the first terminal and the first edge server;

[0083] Based on the transmission distance, energy consumption, and transmission rate of the business data processing model between different edge servers in the preset edge server set, the overall satisfaction of each edge server in the preset edge server set is calculated to obtain the first satisfaction level.

[0084] The highest level of satisfaction is:

[0085]

[0086] in, and These are preset edge servers and centralized edge servers. and The transmission distance and energy consumption of the business data processing model between them. To pre-configure edge servers and centralize edge servers and Between the models, transmission rate To pre-configure edge servers and centralize edge servers Energy consumed in business data processing Pre-set edge servers and centralized edge servers before transmitting business data processing models. Remaining energy.

[0087] Add the third edge server with the highest first satisfaction to the connection with the first terminal.

[0088] It should be noted that in this embodiment, if This indicates that the performance loss is too great, exceeding the processing capacity of a single server, and a single edge server is insufficient to meet the needs of the terminal. For data processing needs, it is necessary to select aggregable edge servers and add them to the edge server collaboration set; the shorter the model transmission time between servers, the lower the energy consumption for server model transmission and computation, indicating that the edge server... The stronger the data processing capability of the edge server, the better. For edge servers Overall satisfaction The higher.

[0089] This embodiment calculates a first satisfaction level. The higher the first satisfaction level, the shorter the model transmission time between servers, and the lower the energy consumption for server model transmission and computation, indicating that the data processing capability of the edge server will be stronger. It can aggregate the business data processing models of multiple edge servers based on the first satisfaction level between edge servers, thereby improving the ability of multiple edge servers to collaboratively process high-frequency acquisition terminal data.

[0090] Overall, the embodiments of the present invention provide an objective standard for evaluating the performance of edge servers by determining the performance loss evaluation threshold of edge servers for power service processing of terminals. By comparing the first loss value with the maximum and minimum loss values, different methods are adopted to select new edge servers or retain existing edge servers under different circumstances. Servers with poor service processing performance can be eliminated, while servers with good service processing performance can be retained. Based on the comprehensive satisfaction among edge servers, the service data processing models of multiple edge servers can be aggregated, improving the ability of multiple edge servers to collaboratively process high-frequency acquisition terminal data. Compared with the prior art, this invention can solve the problem that the prior art cannot effectively improve the collaborative processing capability of multiple edge servers.

[0091] Please see Figure 2 The present invention provides a method for verifying coordination capabilities based on distributed edge computing, comprising:

[0092] S10. According to the above-mentioned collaborative processing method based on distributed edge computing, the terminals in the first data processing model are coordinated and processed, and the connection relationship between each edge server and each terminal in the first data processing model is adjusted to obtain the second data processing model.

[0093] In this embodiment, the edge server collaboration set of the second data processing model is defined as the first collaboration set: .

[0094] S20. Calculate the average synergy index and standard deviation synergy index of the second data processing model, and determine the average synergy threshold and standard deviation synergy threshold. Compare the average synergy index with the average synergy threshold and the standard deviation synergy index with the standard deviation synergy threshold.

[0095] In this embodiment, firstly, according to the first collaborative set of the second data processing model... The transmission latency and computation latency of the business data processing model between various edge servers are used to obtain the average coordination degree index of the coordination set composed of several edge servers connected to several terminals.

[0096] The average synergy index is:

[0097]

[0098] in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

[0099] Then, based on the first collaborative set of the second data processing model The transmission latency and computation latency of business data processing models between various edge servers are calculated, and the standard deviation of the synergy index of the synergy set formed by several edge servers connected to several terminals is calculated.

[0100] The standard deviation coherence index is:

[0101]

[0102] in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

[0103] Secondly, the average synergy threshold is defined as follows: The standard deviation cooperability threshold is ;

[0104] Finally, the average synergy index With average synergy threshold Standard deviation synergy index Coherence threshold with standard deviation Compare them.

[0105] It should be noted that the higher the average collaboration index, the stronger the edge server collaboration set. The lower the average business data processing latency of the edge servers within the second data processing model, the better the overall business processing performance of the edge servers within the model, i.e., the better the aggregation effect. This describes an edge server collaboration set. The degree of fluctuation in processing latency of the inner edge servers; the smaller the fluctuation, the smaller the performance difference between the inner edge servers in the second data processing model.

[0106] This embodiment calculates the average coordination degree index. The larger the average coordination degree index, the smaller the average business data processing latency of the edge servers within the coordination set, and the better the overall business processing performance of the edge servers within the coordination set. It also calculates the standard deviation coordination degree index, which is the reciprocal of the standard deviation of the business data processing latency of the edge server coordination set. This index indicates the degree of fluctuation in the processing latency of the edge servers within the coordination set. The smaller the fluctuation, the smaller the performance difference of the edge servers within the coordination set.

[0107] S30. Validate the second data processing model based on the comparison results until the average synergy index is greater than or equal to the average synergy threshold and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold.

[0108] In this embodiment, if the comparison result is: the average synergy index is greater than or equal to the average synergy threshold, and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold (…), then… ≥ and If the condition is met, the calculation ends.

[0109] If the comparison result is: the average synergy index is less than the average synergy threshold ( < ), or the standard deviation coherence index is less than the standard deviation coherence threshold ( < Then, the scores of each edge server in the second data processing model are recalculated, specifically as follows:

[0110] According to the edge server set in the second data processing model The bandwidth and channel gain between each edge server and the first terminal are calculated, and the score of each edge server in the edge server set by the first terminal is calculated (the calculation principle of this score is the same as the calculation principle of the first score in step S4 of the above-mentioned collaborative processing method based on distributed edge computing).

[0111] Based on the scoring results, disconnect the connection between the fourth edge server with the lowest score in the second data processing model and the terminal connected to it to obtain the third data processing model; replace the second data processing model with the third data processing model, and then perform parameter calculation and comparison on the third data processing model according to the method in step S20. If the comparison result is... ≥ and If the comparison result is..., then the calculation ends; if the comparison result is... < or < Then, the scores of each edge server in the third data processing model are calculated again, and the connection between the edge server and the terminal connected to it is disconnected based on the scores to obtain the fourth data processing model; the process of model substitution and calculation, comparison, and adjustment of the model based on the comparison results is repeated until the average synergy index is greater than or equal to the average synergy threshold and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold. ≥ and ).

[0112] This embodiment controls the average collaboration index to be greater than or equal to the average collaboration threshold, and the standard deviation collaboration index to be greater than or equal to the standard deviation collaboration threshold. This allows for the removal of servers with poor business processing performance and the retention of servers with good business processing performance, thereby improving the overall business processing performance of the edge server collaboration set and reducing the differences in business processing performance among edge servers.

[0113] Overall, by controlling the average collaboration index to be greater than or equal to the average collaboration threshold and the standard deviation collaboration index to be greater than or equal to the standard deviation collaboration threshold, the embodiments of the present invention can improve the overall business processing performance of the edge server collaboration set and reduce the difference in business processing performance between edge servers. Based on a collaborative processing method based on distributed edge computing, it further improves the collaborative processing capability of multiple edge servers.

[0114] Please see Figure 3The present invention provides a collaborative processing device based on distributed edge computing, comprising:

[0115] The first loss value calculation module 10 is used to calculate the first loss value of the power business processing performance of the distribution area based on the first data processing model of distributed edge collaboration, wherein the data processing model is a collaborative set composed of several edge servers connected to the terminal.

[0116] The second loss value determination module 20 is used to determine the power business processing performance loss evaluation threshold of any edge server in the first data processing model for the first terminal connected to it. The upper and lower limits of the threshold are the maximum loss value and the minimum loss value, respectively. The first loss value is compared with the maximum loss value and the minimum loss value.

[0117] The first judgment module 30 is used to maintain the connection between the first terminal and the first edge server if it is determined that the first loss value is less than or equal to the minimum loss value.

[0118] The second judgment module 40 is used to disconnect the connection between the first terminal and the second edge server if it is determined that the first loss value is greater than the minimum loss value and less than the maximum loss value, calculate the score of each edge server in the preset edge server set, and connect the third edge server with the highest score to the first terminal.

[0119] The third judgment module 50 is used to calculate the overall satisfaction of each edge server in the preset edge server set if the first loss value is determined to be greater than or equal to the maximum loss value, and add the fourth edge server with the highest overall satisfaction to the connection with the first terminal.

[0120] This device also includes:

[0121] The coordination capability verification module 60 is used to coordinate each terminal in the first data processing model according to the module described above, adjust the connection relationship between each edge server and each terminal in the first data processing model, and obtain the second data processing model.

[0122] Calculate the average synergy index and standard deviation synergy index of the second data processing model, and determine the average synergy threshold and standard deviation synergy threshold. Compare the average synergy index with the average synergy threshold and the standard deviation synergy index with the standard deviation synergy threshold.

[0123] The second data processing model is validated based on the comparison results until the average synergy index is greater than or equal to the average synergy threshold and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold.

[0124] In one embodiment, the first loss value calculation module 10 is further configured to:

[0125] Based on the data transmission latency, computing resources, and relative load balancing of the distribution terminal, the performance evaluation indicators for power service processing are determined.

[0126] The expected performance index of power business processing is calculated by power business processing performance evaluation index. Based on the power business processing performance evaluation index and the expected performance index of power business processing, the first loss value of power business processing performance of the distribution area is obtained by weighting.

[0127] In one embodiment, the second determining module 40 is further configured to:

[0128] Based on the bandwidth and channel gain between the first terminal and each edge server in the preset edge server set, the score of the first terminal on each edge server in the preset edge server set is calculated, and the first score is obtained. .

[0129] In one embodiment, the third determination module 50 is further configured to:

[0130] Based on the transmission distance, energy consumption, and transmission rate of the business data processing model between different edge servers in the preset edge server set, the overall satisfaction of each edge server in the preset edge server set is calculated to obtain the first satisfaction level.

[0131] The highest level of satisfaction is:

[0132]

[0133] in, and These are preset edge servers and centralized edge servers. and The transmission distance and energy consumption of the business data processing model between them. To pre-configure edge servers and centralize edge servers and Between the models, transmission rate To pre-configure edge servers and centralize edge servers Energy consumed in business data processing Pre-set edge servers and centralized edge servers before transmitting business data processing models. Remaining energy.

[0134] In one embodiment, the coordination capability verification module 60 is further configured to:

[0135] Based on the transmission latency and calculation latency of the business data processing model between each edge server in the first collaborative set of the second data processing model, the average collaboration degree index of the collaborative set composed of several edge servers connected to several terminals is obtained.

[0136] The average synergy index is:

[0137]

[0138] in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

[0139] Based on the transmission latency and computation latency of the business data processing model between each edge server in the first collaborative set of the second data processing model, calculate the standard deviation of the collaborative set formed by several edge servers connected to several terminals.

[0140] The standard deviation coherence index is:

[0141]

[0142] in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

[0143] If the average synergy index is determined to be greater than or equal to the average synergy threshold, and the standard deviation synergy index is determined to be greater than or equal to the standard deviation synergy threshold, then the calculation ends.

[0144] If the average synergy index is determined to be less than the average synergy threshold, or the standard deviation synergy index is less than the standard deviation synergy threshold, then the scores of each edge server in the second data processing model are recalculated. Based on the score results, the connection between the fourth edge server with the lowest score in the second data processing model and the terminal connected to it is disconnected. The second data processing model is updated based on the comparison results until the average synergy index is greater than or equal to the average synergy threshold and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold.

[0145] This device provides an objective standard for evaluating the performance of edge servers by determining a threshold for evaluating the performance loss of power service processing at the terminal. By comparing the first loss value with the maximum and minimum loss values, different methods can be used to select new edge servers or retain existing edge servers under different circumstances. Servers with poor service processing performance can be eliminated, while servers with good service processing performance can be retained. Based on the comprehensive satisfaction among edge servers, the service data processing models of multiple edge servers can be aggregated to improve the ability of multiple edge servers to collaboratively process high-frequency acquisition terminal data. By controlling the average collaboration index to be greater than or equal to the average collaboration threshold and the standard deviation collaboration index to be greater than or equal to the standard deviation collaboration threshold, the overall service processing performance of the edge server collaboration set can be improved, and the differences in service processing performance between edge servers can be reduced. Compared with existing technologies, this device can solve the problem that existing technologies cannot effectively improve the collaborative processing capabilities of multiple edge servers.

[0146] Accordingly, embodiments of the present invention also provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the aforementioned collaborative processing method based on distributed edge computing or the coordination capability verification method based on distributed edge computing;

[0147] The collaborative processing method or coordination capability verification method based on distributed edge computing, if implemented as a software functional unit and used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0148] The above are preferred embodiments of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A collaborative processing method based on distributed edge computing, characterized in that, include: The first loss value of the power service processing performance of the distribution area is calculated based on the first data processing model of distributed edge collaboration. The data processing model is a collaborative set composed of several edge servers connected to the terminal. Specifically, the power service processing performance evaluation index is determined based on the data transmission latency, computing resources, and relative load balance of the edge terminals of the distribution area terminal. The expected performance index of power service processing is calculated through the power service processing performance evaluation index. Based on the power service processing performance evaluation index and the expected performance index of power service processing, the squared difference between the power service processing performance evaluation index and the expected performance index of power service processing is weighted and calculated to obtain the first loss value of the power service processing performance of the distribution area. Determine the power service processing performance loss evaluation threshold for any edge server in the first data processing model to the first terminal connected to it, wherein the upper and lower limits of the threshold are the maximum loss value and the minimum loss value, respectively. If it is determined that the first loss value is less than or equal to the minimum loss value, maintain the connection between the first terminal and the first edge server; If it is determined that the first loss value is greater than the minimum loss value and less than the maximum loss value, disconnect the first terminal from the first edge server, calculate the score of each edge server in the preset edge server set, and connect the second edge server with the highest score to the first terminal. If it is determined that the first loss value is greater than or equal to the maximum loss value, the connection between the first terminal and the first edge server is maintained, the overall satisfaction of each edge server in the preset edge server set is calculated, and the third edge server with the highest overall satisfaction is added to the connection with the first terminal.

2. The collaborative processing method based on distributed edge computing as described in claim 1, characterized in that, Calculate the score of each edge server in the preset edge server set, specifically as follows: Based on the bandwidth and channel gain between the first terminal and each edge server in the preset edge server set, the first terminal calculates the score of each edge server in the preset edge server set to obtain the first score.

3. The collaborative processing method based on distributed edge computing as described in claim 1, characterized in that, The overall satisfaction level of each edge server in the preset edge server set is calculated as follows: Based on the transmission distance, energy consumption, and model transmission rate between different edge servers in the preset edge server set, the overall satisfaction of each edge server in the preset edge server set is calculated to obtain the first satisfaction level. The first satisfaction level is: in, and These are preset edge servers and centralized edge servers. and The transmission distance and energy consumption of the business data processing model between them. To pre-configure edge servers and centralize edge servers and The transmission rate of the business data processing model mentioned above. To pre-configure edge servers and centralize edge servers Energy consumed in business data processing Pre-set edge servers and centralized edge servers before transmitting business data processing models. Remaining energy.

4. A method for verifying coordination capabilities based on distributed edge computing, characterized in that, include: According to any one of claims 1 to 3, the collaborative processing method based on distributed edge computing coordinates the processing of each terminal in the first data processing model, adjusts the connection relationship between each edge server and each terminal in the first data processing model, and obtains the second data processing model. Calculate the average synergy index and standard deviation synergy index of the second data processing model, and determine the average synergy threshold and standard deviation synergy threshold. Compare the average synergy index with the average synergy threshold and the standard deviation synergy index with the standard deviation synergy threshold. The second data processing model is validated based on the comparison results until the average synergy index is greater than or equal to the average synergy threshold and the standard deviation synergy index is greater than or equal to the standard deviation synergy threshold.

5. The coordination capability verification method based on distributed edge computing as described in claim 4, characterized in that, The second data processing model is validated based on the comparison results, specifically as follows: If it is determined that the average synergy index is less than the average synergy threshold, or the standard deviation synergy index is less than the standard deviation synergy threshold, then the scores of each edge server in the second data processing model are recalculated, the connection between the fourth edge server with the lowest score in the second data processing model and the terminal connected to it is disconnected according to the score results, and the second data processing model is updated according to the comparison results.

6. The coordination capability verification method based on distributed edge computing as described in claim 4, characterized in that, The average synergy index of the second data processing model is calculated as follows: Based on the transmission latency and calculation latency of the business data processing model between each edge server in the first collaborative set of the second data processing model, the average collaboration degree index of the collaborative set composed of several edge servers connected to several terminals is obtained. The average degree of synergy index is: in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

7. The coordination capability verification method based on distributed edge computing as described in claim 4, characterized in that, The standard deviation coherence index of the second data processing model is calculated as follows: Based on the transmission latency and computation latency of the business data processing model between each edge server in the first collaborative set of the second data processing model, calculate the standard deviation of the collaborative set formed by the several edge servers connected to several terminals. The standard deviation synergy index is: in, For the first collaborative centralized edge server and Transmission latency of business data processing models between them For the first collaborative centralized edge server and The computational latency of the aggregated business data This represents the total number of edge servers contained in the first collaboration set.

8. A collaborative processing device based on distributed edge computing, characterized in that, include: The first loss value calculation module is used to calculate the first loss value of the power service processing performance of the distribution area based on the first data processing model of distributed edge collaboration. The data processing model is a collaborative set composed of several edge servers connected to the terminal. Specifically, it determines the power service processing performance evaluation index based on the data transmission latency, computing resources, and relative load balance of the edge terminals of the distribution area terminal; calculates the expected performance index of power service processing through the power service processing performance evaluation index; and calculates the first loss value of the power service processing performance by weighting the squared difference between the power service processing performance evaluation index and the expected performance index based on the power service processing performance evaluation index and the expected performance index. The second loss value determination module is used to determine the power service processing performance loss evaluation threshold of any edge server in the first data processing model for the first terminal connected to it. The upper and lower limits of the threshold are the maximum loss value and the minimum loss value, respectively. The first determination module is used to maintain the connection between the first terminal and the first edge server if it is determined that the first loss value is less than or equal to the minimum loss value. The second judgment module is used to disconnect the connection between the first terminal and the second edge server if it is determined that the first loss value is greater than the minimum loss value and less than the maximum loss value, calculate the score of each edge server in the preset edge server set, and connect the third edge server with the highest score to the first terminal. The third judgment module is used to calculate the overall satisfaction of each edge server in the preset edge server set if it is determined that the first loss value is greater than or equal to the maximum loss value, and add the fourth edge server with the highest overall satisfaction to the connection with the first terminal.

9. A storage medium, characterized in that, The storage medium stores a computer program, which is called and executed by a computer to implement the collaborative processing method based on distributed edge computing or the coordination capability verification method based on distributed edge computing as described in any one of claims 1 to 7.

Images