Distributed network flow intelligent scheduling system based on edge computing and AI cooperation

Abstract

The invention relates to the technical field of traffic scheduling, and discloses a distributed network traffic intelligent scheduling system based on edge computing and AI collaboration, which is characterized in that a data acquisition module acquires a first network delay and a first link utilization rate of a distributed network edge node in a preset time period in an ideal state; a second network delay and a second link utilization rate in the actual operation state; the node measurement module determines a node fluctuation measurement value in a preset time period based on the first network delay, the first link utilization rate, the second network delay and the second link utilization rate; the multi-source calculation module analyzes all node fluctuation metric values and calculates node flow blocking coefficients; and the flow scheduling module sets a flow intelligent scheduling strategy of the distributed network edge node according to the node flow blocking coefficient, carries out accurate state measurement on the distributed network edge node, takes multiple indexes and multiple moments as references, deals with the complexity and uncertainty of a modern network, and ensures the scheduling precision and the scheduling intelligence of the network flow.

Description

Technical Field

[0001] This invention relates to the field of traffic scheduling technology, and more specifically, to a distributed network traffic intelligent scheduling system based on edge computing and AI collaboration. Background Technology

[0002] With the rapid development of technologies such as the Internet of Things, 5G communication, industrial internet and streaming media services, the number of network edge devices has exploded, and the resulting data traffic has also shown characteristics of massive volume, real-time processing and low latency.

[0003] Traditional intelligent network traffic scheduling methods are primarily based on traffic scheduling models. These models analyze historical traffic data, learn traffic change patterns, and make intelligent decisions. However, most current traffic scheduling models still rely on centralized training and decision-making in the cloud. Their response speed is limited by communication latency with the cloud, resulting in insufficient real-time decision-making when facing rapid changes in edge networks. Furthermore, existing solutions often focus on optimizing single indicators and single points in time, lacking a comprehensive quantitative evaluation of the overall operation of nodes, leading to insufficiently refined scheduling strategies. Summary of the Invention

[0004] This invention provides a distributed network traffic intelligent scheduling system based on edge computing and AI collaboration. It can accurately measure the state of distributed network edge nodes and optimize multiple indicators and time points to cope with the complexity and uncertainty of modern networks, thereby ensuring the scheduling accuracy and intelligence of network traffic.

[0005] To achieve the above objectives, this invention provides a distributed network traffic intelligent scheduling system based on edge computing and AI collaboration, comprising: The data acquisition module is used to receive network traffic intelligent scheduling commands, pre-set multiple preset time periods, and acquire the first network latency and first link utilization of the distributed network edge nodes under ideal conditions within the preset time periods, and the second network latency and second link utilization under actual operating conditions. Each preset time period includes multiple preset moments. The node measurement module is used to determine the node fluctuation measurement value corresponding to each preset time period based on the first network latency, the first link utilization, the second network latency, and the second link utilization. The multi-source computing module is used to analyze all node fluctuation metrics and calculate the node traffic congestion coefficient of the distributed network edge nodes based on the analysis results. The traffic scheduling module is used to set the intelligent traffic scheduling strategy for the distributed network edge nodes based on the node traffic congestion coefficient.

[0006] Furthermore, it also includes: The command processing module is used for: Obtain the previous command and determine the command category of the previous command; The network traffic intelligent scheduling command is matched with the command category of the previous command, and it is determined whether the network traffic intelligent scheduling command is the same as the previous command. If the network traffic intelligent scheduling command matches the command category of the previous command, then it is determined that the network traffic intelligent scheduling command is the same as the previous command, the first time node of sending the network traffic intelligent scheduling command is collected, and the second time node of sending the previous command is collected. Calculate the time difference between the first time node and the second time node, and determine whether the network traffic intelligent scheduling command can be processed based on the relationship between the time difference and the preset time difference. If the time node difference is greater than or equal to the preset time node difference, it is determined that the network traffic intelligent scheduling command can be processed. If the time node difference is less than the preset time node difference, it is determined that the network traffic intelligent scheduling command cannot be processed, and a log reminder is generated and sent. If the network traffic intelligent scheduling command does not match the command category of the previous command, it is determined that the network traffic intelligent scheduling command is different from the previous command, and the network traffic intelligent scheduling command is processed.

[0007] Furthermore, the node metric module is used for: The node measurement module is used to determine a network delay variation factor based on the first network delay and the second network delay corresponding to each preset time, wherein the network delay variation factor is the absolute value of the difference between the first network delay variation factor and the second network delay variation factor. The node measurement module is used to determine the link utilization change factor based on the first link utilization and the second link utilization corresponding to each preset time, wherein the link utilization change factor is the absolute value of the difference between the first link utilization change factor and the second link utilization change factor. The node measurement module is used to perform a weighted summation of the network latency variation factor and the link utilization variation factor to obtain the node fluctuation measurement value corresponding to a preset time period.

[0008] Furthermore, the node metric module is used for: The node measurement module is used to determine a first network latency value based on the first network latency, wherein the first network latency value is the average of the first network latency within a preset time period; The node measurement module is used to determine multiple first network delay differences based on the first network delay value and the first network delay corresponding to each preset time, wherein the first network delay difference is the absolute value of the difference between the first network delay and the first network delay value; The node metric module is used to obtain the standard deviation of all first network latency differences, which serves as the first network latency variation factor.

[0009] Furthermore, the node metric module is used for: The node measurement module is used to determine the first link utilization value based on the first link utilization, wherein the first link utilization value is the average value of the first link utilization within a preset time period. The node measurement module is used to determine multiple first link utilization differences based on the first link utilization value and the first link utilization corresponding to each preset time, wherein the first link utilization difference is the absolute value of the difference between the first link utilization and the first link utilization value. The node measurement module is used to obtain the standard deviation of all first link utilization differences, which serves as the first link utilization change factor.

[0010] Furthermore, the multi-source computing module is used for: The multi-source computing module is used to determine the smoothness coefficient of the edge node of the distributed network based on all node fluctuation metrics. The multi-source computing module is used to determine the mutation coefficient of the edge node of the distributed network based on all node fluctuation metrics. The multi-source computing module is used to use the sum of the smoothing coefficient and the abrupt change coefficient as the node traffic congestion coefficient of the distributed network edge node.

[0011] Furthermore, the multi-source computing module is used for: The multi-source computing module is used to extract the same node fluctuation metric value from the node fluctuation metric values ​​and obtain multiple node fluctuation metric value sequences. The multi-source computing module is used to count the number of first node fluctuation measurement value sequences in the node fluctuation measurement value sequence; The multi-source computing module is used to extract a node fluctuation metric from each of the node fluctuation metric sequences, and to calculate the first node fluctuation metric and the value. The multi-source computing module is used to calculate the mean of all node fluctuation metric sequences with the same node fluctuation metric value, remove all node fluctuation metric sequences that are less than the mean of the node fluctuation metric sequence, and count the number of second node fluctuation metric sequences in the remaining node fluctuation metric sequences. The multi-source computing module is used to extract a node fluctuation metric from the remaining node fluctuation metric sequence, and to calculate the second node fluctuation metric and the value. The multi-source computing module is used to calculate the smoothness coefficient of the distributed network edge node based on the number of fluctuation metric value sequences of the first node, the number of fluctuation metric value sequences of the second node, the sum of fluctuation metric values ​​of the first node, and the sum of fluctuation metric values ​​of the second node.

[0012] Furthermore, the multi-source computing module is used for: The multi-source computing module is used to calculate the smoothness coefficient of the distributed network edge nodes according to the following formula: ; Where q is the smoothing coefficient of the edge node of the distributed network, n1 is the number of fluctuation metric value sequences of the first node, n2 is the number of fluctuation metric value sequences of the second node, g1 is the sum of fluctuation metric values ​​of the first node, and g2 is the sum of fluctuation metric values ​​of the second node.

[0013] Furthermore, the multi-source computing module is used for: The multi-source computing module is used to extract different node fluctuation metrics; The multi-source computing module is used to calculate the mutation coefficient of the edge nodes of the distributed network according to the following formula: ; Where k is the mutation coefficient of the edge nodes of the distributed network, f is the number of distinct node fluctuation metrics, and y j For the j-th distinct node fluctuation metric, y j+1 This is the fluctuation metric for the (j+1)th distinct node.

[0014] Furthermore, the traffic scheduling module is used for: The traffic scheduling module is used to pre-set a first preset node traffic congestion coefficient and a second preset node traffic congestion coefficient, wherein the first preset node traffic congestion coefficient is less than the second preset node traffic congestion coefficient. The traffic scheduling module is used to generate a data receivable identifier for the distributed network edge node when the node traffic congestion coefficient is less than or equal to the first preset node traffic congestion coefficient. The traffic scheduling module is used to generate a data link maintenance identifier for the distributed network edge node when the node traffic congestion coefficient is greater than the first preset node traffic congestion coefficient and less than the second preset node traffic congestion coefficient. The traffic scheduling module is used to schedule the data of the distributed network edge node to an adjacent edge node when the node traffic congestion coefficient is greater than or equal to the second preset node traffic congestion coefficient, wherein the adjacent edge node carries a data receivable identifier.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention discloses a distributed network traffic intelligent scheduling system based on edge computing and AI collaboration. A data acquisition module acquires the first network latency and first link utilization of distributed network edge nodes under ideal conditions for a preset time period, and the second network latency and second link utilization under actual operating conditions. A node measurement module determines node fluctuation measurement values ​​for the preset time period based on the first network latency, first link utilization, second network latency, and second link utilization. A multi-source computing module analyzes all node fluctuation measurement values ​​and calculates the node traffic congestion coefficient. A traffic scheduling module sets an intelligent traffic scheduling strategy for distributed network edge nodes based on the node traffic congestion coefficient, performing precise state measurement of distributed network edge nodes. Using multiple indicators and multiple time points as benchmarks, this system addresses the complexity and uncertainty of modern networks, ensuring the accuracy and intelligence of network traffic scheduling. Attached Figure Description

[0016] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A schematic diagram of the structure of a distributed network traffic intelligent scheduling system based on edge computing and AI collaboration in an embodiment of the present invention is shown. Detailed Implementation

[0017] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0018] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0019] 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 number of technical features indicated. Therefore, a feature defined as "first" or "second" 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.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] The following is a description of preferred embodiments of the present invention in conjunction with the accompanying drawings.

[0022] like Figure 1 As shown, embodiments of the present invention disclose a distributed network traffic intelligent scheduling system based on edge computing and AI collaboration, including: a data acquisition module, a node measurement module, a multi-source computing module, and a traffic scheduling module.

[0023] In some embodiments of this application, the data acquisition module is used to receive network traffic intelligent scheduling commands, pre-set multiple preset time periods, and acquire the first network latency and first link utilization of the distributed network edge node in the preset time period under ideal conditions, and the second network latency and second link utilization in the actual operating state, wherein each preset time period includes multiple preset moments. In this embodiment, when a network traffic intelligent scheduling command is successfully received, intelligent scheduling of traffic on the distributed network edge nodes begins.

[0024] In this embodiment, the number of preset time periods is preferably 10, and each preset time period includes 10 preset moments. For example, the first preset time period is [second 1, second 10], the second preset time period is [second 11, second 20], and the rest will not be listed one by one.

[0025] In this embodiment, the ideal state refers to the baseline conditions where the network is free from interference, resources are sufficient, and it operates entirely according to theoretical optimality. The actual state refers to the real environment in which the network operates under the influence of dynamic load, external interference, and real-world factors.

[0026] In some embodiments of this application, it also includes: The command processing module is used for: Obtain the previous command and determine the command category of the previous command; The network traffic intelligent scheduling command is matched with the command category of the previous command, and it is determined whether the network traffic intelligent scheduling command is the same as the previous command. If the network traffic intelligent scheduling command matches the command category of the previous command, then it is determined that the network traffic intelligent scheduling command is the same as the previous command, the first time node of sending the network traffic intelligent scheduling command is collected, and the second time node of sending the previous command is collected. Calculate the time difference between the first time node and the second time node, and determine whether the network traffic intelligent scheduling command can be processed based on the relationship between the time difference and the preset time difference. If the time node difference is greater than or equal to the preset time node difference, it is determined that the network traffic intelligent scheduling command can be processed. If the time node difference is less than the preset time node difference, it is determined that the network traffic intelligent scheduling command cannot be processed, and a log reminder is generated and sent. If the network traffic intelligent scheduling command does not match the command category of the previous command, it is determined that the network traffic intelligent scheduling command is different from the previous command, and the network traffic intelligent scheduling command is processed.

[0027] In this embodiment, the preset time difference is preferably 60 minutes, but it can be adjusted adaptively according to actual needs.

[0028] The beneficial effects of the above technical solution are: by setting a preset time node difference, it can effectively avoid repeatedly processing the same or similar intelligent network traffic scheduling commands within a short period of time, thereby reducing the waste of system resources and improving the efficiency and accuracy of command processing. Simultaneously, when it is determined that an intelligent network traffic scheduling command cannot be processed, the system will generate and send a log alert, which helps maintenance personnel to understand the system status in a timely manner, make necessary interventions and adjustments, and ensure the stable operation of the system.

[0029] In some embodiments of this application, the node measurement module is used to determine the node fluctuation measurement value corresponding to each preset time period based on the first network latency, the first link utilization, the second network latency, and the second link utilization. In some embodiments of this application, the node metric module is used for: The node measurement module is used to determine a network delay variation factor based on the first network delay and the second network delay corresponding to each preset time, wherein the network delay variation factor is the absolute value of the difference between the first network delay variation factor and the second network delay variation factor. The node measurement module is used to determine the link utilization change factor based on the first link utilization and the second link utilization corresponding to each preset time, wherein the link utilization change factor is the absolute value of the difference between the first link utilization change factor and the second link utilization change factor. The node measurement module is used to perform a weighted summation of the network latency variation factor and the link utilization variation factor to obtain the node fluctuation measurement value corresponding to a preset time period.

[0030] In this embodiment, the network latency variation factor and the link utilization variation factor are weighted based on the subjective weighting method and the objective weighting method. Here, the weight of the network latency variation factor is preferably 0.65, and the weight of the link utilization variation factor is preferably 0.35.

[0031] The beneficial effects of the above technical solution are: by comprehensively considering network latency variation factors and link utilization variation factors, and performing weighted summation, it can more comprehensively and accurately reflect the fluctuation of distributed network edge nodes within a preset time period. This measurement method not only considers the key indicator of network latency, but also takes into account the dynamic changes in link utilization, thus providing a more reliable and scientific basis for subsequent traffic scheduling.

[0032] In some embodiments of this application, the node metric module is used for: The node measurement module is used to determine a first network latency value based on the first network latency, wherein the first network latency value is the average of the first network latency within a preset time period; The node measurement module is used to determine multiple first network delay differences based on the first network delay value and the first network delay corresponding to each preset time, wherein the first network delay difference is the absolute value of the difference between the first network delay and the first network delay value; The node metric module is used to obtain the standard deviation of all first network latency differences, which serves as the first network latency variation factor.

[0033] The beneficial effects of the above technical solution are: by calculating the mean of the first network latency and the absolute value of the difference between each preset time point and the mean, and further calculating the standard deviation of these absolute values ​​of difference, the fluctuation of network latency within a preset time period can be quantified more accurately. This quantification method helps to more accurately assess the stability of network latency and provides more detailed and accurate data support for subsequent traffic scheduling.

[0034] In some embodiments of this application, the node metric module is used for: The node measurement module is used to determine a second network latency value based on the second network latency, wherein the second network latency value is the average value of the second network latency within a preset time period; The node measurement module is used to determine multiple second network delay differences based on the second network delay value and the second network delay corresponding to each preset time, wherein the second network delay difference is the absolute value of the difference between the second network delay and the second network delay value; The node metric module is used to obtain the standard deviation of all second network latency differences, which serves as the second network latency variation factor.

[0035] In some embodiments of this application, the node metric module is used for: The node measurement module is used to determine the first link utilization value based on the first link utilization, wherein the first link utilization value is the average value of the first link utilization within a preset time period. The node measurement module is used to determine multiple first link utilization differences based on the first link utilization value and the first link utilization corresponding to each preset time, wherein the first link utilization difference is the absolute value of the difference between the first link utilization and the first link utilization value. The node measurement module is used to obtain the standard deviation of all first link utilization differences, which serves as the first link utilization change factor.

[0036] The beneficial effects of the above technical solution are as follows: by calculating the mean of the first link utilization rate and the absolute value of the difference between the mean and the mean at each preset time point, and further calculating the standard deviation of these absolute values ​​of difference, the fluctuation of the link utilization rate within the preset time period can be quantified more accurately. This quantification method helps to more accurately assess the stability of the link utilization rate, providing more detailed and accurate data support for subsequent traffic scheduling, ensuring that the scheduling strategy can fully consider the dynamic changes in link utilization rate, thereby improving the scheduling efficiency and stability of network traffic.

[0037] In some embodiments of this application, the node metric module is used for: The node measurement module is used to determine the second link utilization value based on the second link utilization rate, wherein the second link utilization rate value is the average value of the second link utilization rate within a preset time period; The node measurement module is used to determine multiple second link utilization differences based on the second link utilization value and the second link utilization corresponding to each preset time, wherein the second link utilization difference is the absolute value of the difference between the second link utilization and the second link utilization value; The node metric module is used to obtain the standard deviation of all second link utilization differences, which serves as the second link utilization variation factor.

[0038] In some embodiments of this application, a multi-source computing module is used to analyze all node fluctuation metrics and calculate the node traffic congestion coefficient of the distributed network edge node based on the analysis results. In some embodiments of this application, the multi-source computing module is used for: The multi-source computing module is used to determine the smoothness coefficient of the edge node of the distributed network based on all node fluctuation metrics. The multi-source computing module is used to determine the mutation coefficient of the edge node of the distributed network based on all node fluctuation metrics. The multi-source computing module is used to use the sum of the smoothing coefficient and the abrupt change coefficient as the node traffic congestion coefficient of the distributed network edge node.

[0039] The beneficial effects of the above technical solution are as follows: by calculating the smoothness coefficient and the abrupt change coefficient separately, and using their sum as the node traffic congestion coefficient, the traffic congestion status of edge nodes in a distributed network can be reflected more comprehensively. The smoothness coefficient reflects the stability of node traffic changes over a longer period, while the abrupt change coefficient can capture sudden changes in node traffic. This comprehensive approach helps to more accurately assess the traffic congestion risk of nodes, providing a more scientific and reasonable basis for subsequent traffic scheduling, thereby improving the overall scheduling effect and stability of network traffic.

[0040] In some embodiments of this application, the multi-source computing module is used for: The multi-source computing module is used to extract the same node fluctuation metric value from the node fluctuation metric values ​​and obtain multiple node fluctuation metric value sequences. The multi-source computing module is used to count the number of first node fluctuation measurement value sequences in the node fluctuation measurement value sequence; The multi-source computing module is used to extract a node fluctuation metric from each of the node fluctuation metric sequences, and to calculate the first node fluctuation metric and the value. The multi-source computing module is used to calculate the mean of all node fluctuation metric sequences with the same node fluctuation metric value, remove all node fluctuation metric sequences that are less than the mean of the node fluctuation metric sequence, and count the number of second node fluctuation metric sequences in the remaining node fluctuation metric sequences. The multi-source computing module is used to extract a node fluctuation metric from the remaining node fluctuation metric sequence, and to calculate the second node fluctuation metric and the value. The multi-source computing module is used to calculate the smoothness coefficient of the distributed network edge node based on the number of fluctuation metric value sequences of the first node, the number of fluctuation metric value sequences of the second node, the sum of fluctuation metric values ​​of the first node, and the sum of fluctuation metric values ​​of the second node.

[0041] In this embodiment, the node fluctuation metric values ​​in each node fluctuation metric value sequence are the same, and the number of node fluctuation metric values ​​in each node fluctuation metric value sequence is greater than or equal to 2. The node fluctuation metric values ​​are different between different node fluctuation metric value sequences.

[0042] The beneficial effects of the above technical solution are: by calculating the smoothness coefficient of the edge node of the distributed network based on the number of fluctuation metric sequence of the first node, the number of fluctuation metric sequence of the second node, the sum of fluctuation metric values ​​of the first node and the sum of fluctuation metric values ​​of the second node, it is possible to more accurately measure the smoothness of traffic changes of the node over a longer period of time.

[0043] In some embodiments of this application, the multi-source computing module is used for: The multi-source computing module is used to calculate the smoothness coefficient of the distributed network edge nodes according to the following formula: ; Where q is the smoothing coefficient of the edge node of the distributed network, n1 is the number of fluctuation metric value sequences of the first node, n2 is the number of fluctuation metric value sequences of the second node, g1 is the sum of fluctuation metric values ​​of the first node, and g2 is the sum of fluctuation metric values ​​of the second node.

[0044] In some embodiments of this application, the multi-source computing module is used for: The multi-source computing module is used to extract different node fluctuation metrics; The multi-source computing module is used to calculate the mutation coefficient of the edge nodes of the distributed network according to the following formula: ; Where k is the mutation coefficient of the edge nodes of the distributed network, f is the number of distinct node fluctuation metrics, and y j For the j-th distinct node fluctuation metric, y j+1 This is the fluctuation metric for the (j+1)th distinct node.

[0045] In this embodiment, the number of different node fluctuation metrics is one.

[0046] The beneficial effect of the above technical solution is that it can accurately capture sudden changes in node traffic.

[0047] In some embodiments of this application, the traffic scheduling module is used to set the intelligent traffic scheduling strategy for the distributed network edge nodes based on the node traffic congestion coefficient.

[0048] In some embodiments of this application, the traffic scheduling module is used for: The traffic scheduling module is used to pre-set a first preset node traffic congestion coefficient and a second preset node traffic congestion coefficient, wherein the first preset node traffic congestion coefficient is less than the second preset node traffic congestion coefficient. The traffic scheduling module is used to generate a data receivable identifier for the distributed network edge node when the node traffic congestion coefficient is less than or equal to the first preset node traffic congestion coefficient. The traffic scheduling module is used to generate a data link maintenance identifier for the distributed network edge node when the node traffic congestion coefficient is greater than the first preset node traffic congestion coefficient and less than the second preset node traffic congestion coefficient. The traffic scheduling module is used to schedule the data of the distributed network edge node to an adjacent edge node when the node traffic congestion coefficient is greater than or equal to the second preset node traffic congestion coefficient, wherein the adjacent edge node carries a data receivable identifier.

[0049] In this embodiment, the preferred flow congestion coefficient of the first preset node is 3, and the preferred flow congestion coefficient of the second preset node is 6. The specific values ​​can be adjusted according to actual needs.

[0050] The beneficial effects of the above technical solution are as follows: By pre-setting a first preset node traffic congestion coefficient and a second preset node traffic congestion coefficient, and based on the comparison results between the node traffic congestion coefficient and these two preset values, data receptivity identifiers, data link maintenance identifiers, or data scheduling can be flexibly generated, enabling intelligent and efficient management of traffic at the edge nodes of the distributed network. This hierarchical processing strategy not only ensures data reception efficiency during low-traffic congestion but also maintains data link stability during medium-traffic congestion. Furthermore, it promptly schedules data to adjacent available nodes during high-traffic congestion, thereby effectively avoiding network congestion and data loss, and improving the operational efficiency and reliability of the entire distributed network system.

[0051] In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0052] Although the invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the embodiments disclosed in this invention can be combined with each other in any way. The fact that not all of these combinations are described in this specification is merely for the sake of brevity and resource conservation.

[0053] It will be understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A distributed network traffic intelligent scheduling system based on edge computing and AI collaboration, characterized in that, include: The data acquisition module is used to receive network traffic intelligent scheduling commands, pre-set multiple preset time periods, and acquire the first network latency and first link utilization of the distributed network edge nodes under ideal conditions within the preset time periods, and the second network latency and second link utilization under actual operating conditions. Each preset time period includes multiple preset moments. The node measurement module is used to determine the node fluctuation measurement value corresponding to each preset time period based on the first network latency, the first link utilization, the second network latency, and the second link utilization. The multi-source computing module is used to analyze all node fluctuation metrics and calculate the node traffic congestion coefficient of the distributed network edge nodes based on the analysis results. The traffic scheduling module is used to set the intelligent traffic scheduling strategy for the distributed network edge nodes based on the node traffic congestion coefficient.

2. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 1, characterized in that, Also includes: The command processing module is used for: Obtain the previous command and determine the command category of the previous command; The network traffic intelligent scheduling command is matched with the command category of the previous command, and it is determined whether the network traffic intelligent scheduling command is the same as the previous command. If the network traffic intelligent scheduling command matches the command category of the previous command, then it is determined that the network traffic intelligent scheduling command is the same as the previous command, the first time node of sending the network traffic intelligent scheduling command is collected, and the second time node of sending the previous command is collected. Calculate the time difference between the first time node and the second time node, and determine whether the network traffic intelligent scheduling command can be processed based on the relationship between the time difference and the preset time difference. If the time node difference is greater than or equal to the preset time node difference, it is determined that the network traffic intelligent scheduling command can be processed. If the time node difference is less than the preset time node difference, it is determined that the network traffic intelligent scheduling command cannot be processed, and a log reminder is generated and sent. If the network traffic intelligent scheduling command does not match the command category of the previous command, it is determined that the network traffic intelligent scheduling command is different from the previous command, and the network traffic intelligent scheduling command is processed.

3. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 1, characterized in that, The node metric module is used for: The node measurement module is used to determine a network delay variation factor based on the first network delay and the second network delay corresponding to each preset time, wherein the network delay variation factor is the absolute value of the difference between the first network delay variation factor and the second network delay variation factor. The node measurement module is used to determine the link utilization change factor based on the first link utilization and the second link utilization corresponding to each preset time, wherein the link utilization change factor is the absolute value of the difference between the first link utilization change factor and the second link utilization change factor. The node measurement module is used to perform a weighted summation of the network latency variation factor and the link utilization variation factor to obtain the node fluctuation measurement value corresponding to a preset time period.

4. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 3, characterized in that, The node metric module is used for: The node measurement module is used to determine a first network latency value based on the first network latency, wherein the first network latency value is the average of the first network latency within a preset time period; The node measurement module is used to determine multiple first network delay differences based on the first network delay value and the first network delay corresponding to each preset time, wherein the first network delay difference is the absolute value of the difference between the first network delay and the first network delay value; The node metric module is used to obtain the standard deviation of all first network latency differences, which serves as the first network latency variation factor.

5. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 3, characterized in that, The node metric module is used for: The node measurement module is used to determine the first link utilization value based on the first link utilization, wherein the first link utilization value is the average value of the first link utilization within a preset time period. The node measurement module is used to determine multiple first link utilization differences based on the first link utilization value and the first link utilization corresponding to each preset time, wherein the first link utilization difference is the absolute value of the difference between the first link utilization and the first link utilization value. The node measurement module is used to obtain the standard deviation of all first link utilization differences, which serves as the first link utilization change factor.

6. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 1, characterized in that, The multi-source computing module is used for: The multi-source computing module is used to determine the smoothness coefficient of the edge node of the distributed network based on all node fluctuation metrics. The multi-source computing module is used to determine the mutation coefficient of the edge node of the distributed network based on all node fluctuation metrics. The multi-source computing module is used to use the sum of the smoothing coefficient and the abrupt change coefficient as the node traffic congestion coefficient of the distributed network edge node.

7. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 6, characterized in that, The multi-source computing module is used for: The multi-source computing module is used to extract the same node fluctuation metric value from the node fluctuation metric values ​​and obtain multiple node fluctuation metric value sequences. The multi-source computing module is used to count the number of first node fluctuation measurement value sequences in the node fluctuation measurement value sequence; The multi-source computing module is used to extract a node fluctuation metric from each of the node fluctuation metric sequences, and to calculate the first node fluctuation metric and the value. The multi-source computing module is used to calculate the mean of all node fluctuation metric sequences with the same node fluctuation metric value, remove all node fluctuation metric sequences that are less than the mean of the node fluctuation metric sequence, and count the number of second node fluctuation metric sequences in the remaining node fluctuation metric sequences. The multi-source computing module is used to extract a node fluctuation metric from the remaining node fluctuation metric sequence, and to calculate the second node fluctuation metric and the value. The multi-source computing module is used to calculate the smoothness coefficient of the distributed network edge node based on the number of fluctuation metric value sequences of the first node, the number of fluctuation metric value sequences of the second node, the sum of fluctuation metric values ​​of the first node, and the sum of fluctuation metric values ​​of the second node.

8. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 7, characterized in that, The multi-source computing module is used for: The multi-source computing module is used to calculate the smoothness coefficient of the distributed network edge nodes according to the following formula: ； Where q is the smoothing coefficient of the edge node of the distributed network, n1 is the number of fluctuation metric value sequences of the first node, n2 is the number of fluctuation metric value sequences of the second node, g1 is the sum of fluctuation metric values ​​of the first node, and g2 is the sum of fluctuation metric values ​​of the second node.

9. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 6, characterized in that, The multi-source computing module is used for: The multi-source computing module is used to extract different node fluctuation metrics; The multi-source computing module is used to calculate the mutation coefficient of the edge nodes of the distributed network according to the following formula: ； Where k is the mutation coefficient of the edge nodes of the distributed network, f is the number of distinct node fluctuation metrics, and y j For the j-th distinct node fluctuation metric, y j+1 This is the fluctuation metric for the (j+1)th distinct node.

10. The distributed network traffic intelligent scheduling system based on edge computing and AI collaboration according to claim 1, characterized in that, The traffic scheduling module is used for: The traffic scheduling module is used to pre-set a first preset node traffic congestion coefficient and a second preset node traffic congestion coefficient, wherein the first preset node traffic congestion coefficient is less than the second preset node traffic congestion coefficient. The traffic scheduling module is used to generate a data receivable identifier for the distributed network edge node when the node traffic congestion coefficient is less than or equal to the first preset node traffic congestion coefficient. The traffic scheduling module is used to generate a data link maintenance identifier for the distributed network edge node when the node traffic congestion coefficient is greater than the first preset node traffic congestion coefficient and less than the second preset node traffic congestion coefficient. The traffic scheduling module is used to schedule the data of the distributed network edge node to an adjacent edge node when the node traffic congestion coefficient is greater than or equal to the second preset node traffic congestion coefficient, wherein the adjacent edge node carries a data receivable identifier.

Images