Method, system and equipment for processing packet loss of available area network on cloud and medium

Abstract

The invention provides an on-cloud available area network packet loss processing method, system and device and a medium, and the method specifically comprises the steps: respectively deploying a plurality of detection nodes on each on-cloud available area network, and obtaining the multi-dimensional network packet loss index data of each on-cloud available area network according to the detection nodes; processing the multi-dimensional network packet loss index data of all the detection nodes of each on-cloud available area network by adopting a weighted fusion algorithm to obtain a network packet loss comprehensive score of each on-cloud available area network; and determining a network packet loss score threshold value of each on-cloud available area network according to the network area division, the service type and the time period configuration parameters, and determining whether a network packet loss fault event occurs in each on-cloud available area network by comparing the network packet loss comprehensive score and the network packet loss score threshold value of each on-cloud available area network. The packet loss monitoring accuracy and comprehensiveness of the complex network in the cloud environment are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of network communication technology, and in particular to a method, system, device and medium for processing network packet loss in an available zone on a cloud. Background Art

[0002] With the rapid development of cloud computing technology, more and more companies are choosing to migrate their businesses to the cloud to achieve flexible configuration and efficient use of resources. However, while enjoying the convenience brought by cloud computing, the stability and reliability of cloud networks are becoming increasingly prominent, among which the problem of network packet loss is particularly prominent. Network packet loss refers to the failure of some data packets to be successfully transmitted to the destination during network transmission due to network congestion, transmission errors, equipment failure and other reasons, thereby affecting the integrity and accuracy of data transmission.

[0003] Network packet loss not only reduces the efficiency of data transmission, but can also cause business interruption and data loss, bringing immeasurable losses to enterprises. In the cloud environment, due to the complex network structure, numerous and widely distributed devices, it is more difficult to find and locate network packet loss problems. Traditional network monitoring and troubleshooting methods often rely on manual intervention, which is inefficient and prone to errors, and cannot meet the high reliability and high availability requirements of cloud services.

[0004] Therefore, how to monitor the network status of cloud availability zones in real time and accurately, and promptly detect and handle network packet loss issues, has become a technical problem that needs to be solved in the current cloud computing field. At present, although there are some network monitoring and troubleshooting tools on the market, most of them have problems such as single monitoring indicators, low judgment accuracy, and insufficient automation, which makes it difficult to meet the actual needs of cloud network management. Summary of the invention

[0005] The purpose of the present invention is to provide a method, system, device and medium for processing packet loss in a cloud availability zone network, which effectively improves the accuracy and comprehensiveness of packet loss monitoring in complex networks in a cloud environment, realizes accurate fault location and personalized notification, and significantly improves the efficiency of network fault processing, so as to solve at least one of the above-mentioned prior art problems.

[0006] In a first aspect, the present invention provides a method for processing network packet loss in an available zone on a cloud, the method specifically comprising:

[0007] Deploy several detection nodes in each available zone network on each cloud, and obtain multi-dimensional network packet loss indicator data of each available zone network on each cloud according to the detection nodes;

[0008] A weighted fusion algorithm is used to process the multi-dimensional network packet loss indicator data of all detection nodes in each cloud availability zone network to obtain a comprehensive network packet loss score for each cloud availability zone network.

[0009] Determine the network packet loss score threshold of each cloud availability zone network based on network area division, service type, and time period configuration parameters. Determine whether a network packet loss failure event has occurred in each cloud availability zone network by comparing the network packet loss comprehensive score of each cloud availability zone network with the network packet loss score threshold.

[0010] Determine packet loss failure event data of a network in an available zone on the cloud where a network packet loss failure event has occurred, and determine a packet loss event level according to the packet loss failure event data;

[0011] Based on the packet loss event level and the attribute information of each operation and maintenance personnel, different network fault briefing generation rules and different network fault briefing push rules are determined for each operation and maintenance personnel, and corresponding network fault briefings are generated and pushed for each operation and maintenance personnel according to the different network fault briefing generation rules and different network fault briefing push rules.

[0012] In a second aspect, the present invention provides a system for processing network packet loss in an available zone on a cloud, the system specifically comprising:

[0013] A first processing module is used to deploy a plurality of detection nodes in each available zone network on each cloud, and obtain multi-dimensional network packet loss indicator data of each available zone network on each cloud according to the detection nodes;

[0014] The second processing module is used to use a weighted fusion algorithm to process the multi-dimensional network packet loss indicator data of all detection nodes of each cloud available zone network, and obtain a comprehensive network packet loss score for each cloud available zone network;

[0015] The third processing module is used to determine the network packet loss score threshold of each cloud availability zone network according to the network area division, service type and time period configuration parameters, and determine whether a network packet loss failure event has occurred in each cloud availability zone network by comparing the network packet loss comprehensive score of each cloud availability zone network with the network packet loss score threshold;

[0016] A fourth processing module is used to determine packet loss failure event data of the cloud available zone network where the network packet loss failure event has occurred, and determine the packet loss event level according to the packet loss failure event data;

[0017] The fifth processing module is used to determine different network fault briefing generation rules and different network fault briefing push rules for each operation and maintenance personnel based on the packet loss event level and the attribute information of each operation and maintenance personnel, and generate and push corresponding network fault briefings for each operation and maintenance personnel according to different network fault briefing generation rules and different network fault briefing push rules.

[0018] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor and a computer program stored in the memory, and when the computer program is executed on the processor, a method for processing network packet loss in an available zone on the cloud as described in any one of the above methods is implemented.

[0019] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for processing network packet loss in an available zone on the cloud as described in any one of the above methods is implemented.

[0020] Compared with the prior art, the present invention has at least one of the following technical effects:

[0021] 1. The present invention effectively solves the problem of network packet loss in the cloud availability zone through real-time monitoring, accurate judgment, flexible configuration and automatic processing, ensures business continuity and data security, and reduces the losses suffered by enterprises due to network packet loss.

[0022] 2. The present invention can intelligently detect network packet loss failures in a multi-cloud environment, and provide personalized network fault briefings to different operation and maintenance personnel according to the severity and impact scope of the fault, thereby improving the timeliness and effectiveness of network fault handling and ensuring the stable operation of the business.

[0023] 3. The present invention effectively improves the accuracy and comprehensiveness of packet loss monitoring in complex networks in cloud environments, achieves accurate fault location and personalized notification, and significantly improves the efficiency of network fault handling.

[0024] 4. The present invention deploys detection nodes in the network of each available zone on the cloud to obtain multi-dimensional network packet loss indicator data in real time, thereby realizing real-time monitoring of the network status on the cloud and promptly discovering packet loss problems.

[0025] 5. The present invention uses a weighted fusion algorithm to process the multi-dimensional network packet loss index data of the detection node to obtain a comprehensive network packet loss score, and determines the packet loss score threshold in combination with configuration parameters such as network area division, service type and time period, which can accurately determine whether packet loss occurs and its severity.

[0026] 6. The present invention supports customizing network packet loss scenarios and setting packet loss thresholds, adapts to network environments of different scales and complexities, supports multiple cloud platforms, and improves the flexibility and applicability of the method.

[0027] 7. The present invention automatically generates and pushes personalized network fault briefings according to the packet loss event level and the attribute information of the operation and maintenance personnel, realizes the automatic determination and processing of network packet loss problems, and improves the efficiency of network operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 It is a flowchart of a method for processing network packet loss in an available zone on a cloud provided by the first embodiment of the present invention;

[0030] Figure 2 It is a flowchart of a method for processing network packet loss in an available zone on a cloud provided by the second embodiment of the present invention;

[0031] Figure 3 It is a structural diagram of a system for processing network packet loss in an available zone on a cloud provided by an embodiment of the present invention;

[0032] Figure 4 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0034] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0035] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0036] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0037] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0038] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0039] In the embodiment of the present application, the execution subject of the process includes a terminal device, which includes but is not limited to: a server, a computer, a smart phone, a tablet computer, and other devices capable of executing the method disclosed in the present application. Figure 1 A flowchart of a method for processing packet loss in a cloud availability zone network disclosed in the first embodiment of the present invention is shown, and the details are as follows:

[0040] S101, deploying a plurality of detection nodes in the available zone networks on each cloud, and obtaining multi-dimensional network packet loss indicator data of the available zone networks on each cloud according to the detection nodes.

[0041] In this embodiment, in order to monitor the multi-dimensional network packet loss indicator data of each available zone network on the cloud, a network monitoring system based on detection nodes is designed and deployed. The system collects and analyzes key performance indicators such as network packet loss, latency, and jitter in real time by deploying several detection nodes in each available zone on the cloud, providing data support for network optimization and troubleshooting.

[0042] Specifically, according to the network architecture and scale of the cloud availability zone, select appropriate servers or virtual machines as detection nodes. These nodes should have stable network connections and sufficient computing resources. Deploy detection nodes near key network nodes in each cloud availability zone to ensure that the network status of the availability zone can be fully covered. Install network monitoring software on the detection nodes, such as Ping, Traceroute, iperf and other tools, as well as customized network monitoring scripts to send and receive network data packets and record relevant indicator data.

[0043] Configure the detection node to regularly send ICMP protocol packets, TCP / UDP data packets, etc. to the target server and its intermediate routing nodes, and collect multi-dimensional network performance indicator data including packet loss rate, data delay, data jitter, available bandwidth estimation, routing hop count, etc. Use cloud network analysis technology to analyze the collected network data in real time to identify abnormal situations such as network congestion, delay, packet loss, etc. Through visualization tools, the analysis results are displayed in the form of charts, trend charts, etc., to help operation and maintenance personnel intuitively understand the network status. Based on the analysis results, quickly locate network fault points, such as network equipment failure, line failure, etc. According to the network performance bottleneck, make optimization suggestions, such as adjusting the network topology, increasing bandwidth, optimizing applications, etc.

[0044] In this embodiment, by real-time monitoring and analysis of key indicators such as network packet loss, network problems are discovered and solved in a timely manner to improve the stability and reliability of the network. Based on the monitoring data, the network is finely managed and optimized, such as adjusting the network bandwidth, optimizing the routing strategy, etc., to improve the overall performance of the network.

[0045] In some embodiments, in the above step S101, the obtaining of network performance data of each available zone network on the cloud according to the detection node also includes:

[0046] Obtain the topology information and path dynamic change information of each available zone network on the cloud, and determine the initial detection path and detection frequency parameters according to the topology information and the path dynamic change information;

[0047] The ARIMA time series analysis algorithm is used to model and analyze the historical detection data of the network in each available zone on the cloud to obtain the probability distribution of packet loss under different paths and frequencies.

[0048] According to the initial detection path, the detection frequency parameters and the packet loss probability distribution under different paths and at different frequencies, an integer programming detection strategy optimization model is constructed with the minimization of the overall packet loss monitoring blind area as the optimization goal. The integer programming detection strategy optimization model is used to determine the optimal detection strategy for each available zone network on the cloud.

[0049] In this embodiment, the network topology information of the available zone on the cloud is obtained, including network nodes, links, and the connection relationship between them, etc., to build a network topology graph model. The dynamic change information of each path in the available zone network on the cloud is monitored, including the changes in performance indicators such as path delay, packet loss rate, bandwidth, etc., and the path dynamic change data is recorded. According to the network topology information and the dynamic change information of the path, the graph theory algorithm and the machine learning algorithm are used to automatically plan and generate the initial network detection path, covering the key paths and nodes. By analyzing the statistical characteristics of the dynamic change data of the path, the clustering algorithm is used to group the paths to obtain stable path groups and unstable path groups. For different path groups, the detection frequency parameters are adaptively set, the stable path group has a low detection frequency, the unstable path group has a high detection frequency, and the detection efficiency is dynamically optimized.

[0050] Obtain the historical network detection data of the available zones on the cloud, and classify and organize the data under different network paths and detection frequencies. Use the ARIMA time series analysis algorithm to model and analyze the classified historical detection data. According to the results of the ARIMA algorithm modeling analysis, the packet loss probability distribution under different network paths and detection frequencies is obtained. If the packet loss probability under a certain network path or detection frequency exceeds the preset threshold, it is judged that the path or frequency has a network anomaly risk. For paths or frequencies with network anomaly risks, the exponential smoothing method is used to predict the packet loss probability in the future.

[0051] The initial detection path, detection frequency parameters, and packet loss probability distribution data under different paths and frequencies are obtained as inputs for constructing an integer programming detection strategy optimization model. Based on the obtained data, the detection strategy optimization model is constructed using the integer programming algorithm, and the objective function is set to minimize the overall packet loss monitoring blind area. In the constructed integer programming detection strategy optimization model, the decision variables are set as the detection path and detection frequency of each cloud availability zone network. By solving the integer programming detection strategy optimization model, the optimal detection path and detection frequency combination of each cloud availability zone network, that is, the optimal detection strategy, is obtained. If the solved optimal detection strategy can meet the goal of minimizing the overall packet loss monitoring blind area, the strategy is applied to the packet loss monitoring of each cloud availability zone network. After applying the optimal detection strategy, the packet loss probability distribution data of each cloud availability zone network is continuously obtained for dynamically updating the integer programming detection strategy optimization model. According to the updated integer programming detection strategy optimization model, the optimal detection strategy is periodically re-solved and applied to the cloud availability zone network to adapt to changes in network conditions and continuously minimize the overall packet loss monitoring blind area.

[0052] Exemplarily, in order to obtain the topological structure information and path dynamic change information of each cloud available zone network, a network topology discovery tool such as Nmap can be used to scan the target network, obtain the device information, link information and routing information in the network, and at the same time, by continuously monitoring the changes in network traffic, a machine learning algorithm such as a support vector machine SVM is used to learn and predict the path change law, so as to grasp the network topology dynamics in real time. Based on the above information, the initial detection path can be constructed by the minimum spanning tree algorithm, with the principle of covering the entire network and the shortest path, and the initial detection frequency is set to once every 5 minutes according to factors such as network bandwidth and device load. After obtaining sufficient historical detection data, the packet loss rate of each path is modeled using the ARIMA time series analysis algorithm, and the packet loss probability distribution under different paths and frequencies is obtained by parameter estimation. Finally, with the goal of comprehensive path coverage and the lowest total packet loss rate, an integer programming model is constructed to solve how to select the optimal detection path set and the detection frequency of each path, so that the blind spot of the whole network packet loss monitoring is minimized when the monitoring resources are limited. Through the above technical solution, the detection strategy can be adaptively optimized according to the dynamic characteristics of the cloud network, thereby reducing overhead while ensuring the monitoring quality and improving the reliability of the cloud network.

[0053] S102, using a weighted fusion algorithm, respectively processes the multi-dimensional network packet loss indicator data of all detection nodes of each cloud availability zone network to obtain a comprehensive network packet loss score for each cloud availability zone network.

[0054] In this embodiment, in order to more accurately evaluate the network quality of each cloud availability zone network, especially the network packet loss situation, a network packet loss comprehensive scoring system based on a weighted fusion algorithm is designed and implemented. The system performs weighted fusion processing on the multi-dimensional network packet loss indicator data of all detection nodes in each cloud availability zone network, and finally obtains a comprehensive score to quantify the network packet loss situation.

[0055] Specifically, first determine the weight distribution principle of the weighted fusion algorithm. The weight can be distributed according to factors such as the location, importance, and historical performance of the detection node. For example, the detection node located on the critical path of the network can have a higher weight. For each cloud available zone network, the multi-dimensional network packet loss indicator data of all its internal detection nodes is used as input. According to the weight distribution principle, the data of each detection node is weighted. The weighted data is fused to calculate the comprehensive network packet loss score of the cloud available zone network. The score can be expressed in percentage or other forms for easy understanding and comparison. The calculated comprehensive network packet loss score is displayed in a visual form to relevant personnel, such as operation and maintenance personnel, management personnel, etc. Provide detailed scoring reports and data analysis to help relevant personnel understand the network status and formulate corresponding optimization measures.

[0056] In this embodiment, the multi-dimensional data of multiple detection nodes are comprehensively processed by the weighted fusion algorithm, which can more comprehensively reflect the network packet loss status and improve the accuracy of the evaluation. Considering that the data of different detection nodes may have differences or errors, the weighted fusion algorithm can flexibly process the data according to the weight distribution principle, reduce the impact of individual abnormal data on the overall evaluation results, and enhance the robustness of the system.

[0057] In some embodiments, in the above step S102, the weighted fusion algorithm is used to process the multi-dimensional network packet loss index data of all detection nodes of each cloud available zone network respectively to obtain a comprehensive network packet loss score of each cloud available zone network, specifically including:

[0058] The hierarchical analysis method is used to determine the weight coefficient of the network packet loss index data of each detection node;

[0059] According to the preset weighted fusion algorithm model and the weight coefficient, the network packet loss index data of each detection node is weightedly calculated to obtain a comprehensive network packet loss score for each detection node;

[0060] The K-means clustering algorithm is used to perform cluster analysis on the comprehensive scores of network packet loss of all detection nodes to determine the network packet loss level data of each detection node;

[0061] Obtain network packet loss level data of all detection nodes of each available zone network on each cloud, and obtain the distribution of network packet loss levels of detection nodes of each available zone network on each cloud by performing frequency statistics on the network packet loss level data;

[0062] According to the preset network packet loss level weight coefficient, the distribution of the network packet loss level of the detection node is weightedly calculated to obtain a comprehensive network packet loss score for each available zone network on the cloud.

[0063] In this embodiment, network packet loss index data collected by multiple detection nodes, including packet loss rate, delay, jitter and other indicators, are obtained as input of the hierarchical analysis method. Through the hierarchical analysis method, a hierarchical structure model of network packet loss indicators is constructed, the importance weights between the indicators are determined, and the weight coefficients of the network packet loss index data of each detection node are calculated. According to the preset weighted fusion algorithm model, the weight coefficients are weightedly calculated with the network packet loss index data of the corresponding detection node to obtain the comprehensive score of network packet loss of each detection node.

[0064] The K-means clustering algorithm is used to take the network packet loss scores of all detection nodes as input, and the nodes are divided into different clusters through iterative optimization. Each cluster represents a network packet loss level. In the clustering process, the distance between the node score and the cluster center is calculated, and the node is divided into the cluster with the closest distance. The cluster center is continuously updated until the clustering results converge. According to the clustering results, the network packet loss level of each detection node is determined, and the network quality classification label of the node is obtained.

[0065] Obtain all detection node information in each available zone on the cloud, including metadata such as node ID and available zone where the node is located, and store it in the detection node information table. For each detection node, periodically perform network packet loss tests to obtain network packet loss rate data over a period of time. According to the preset packet loss rate level threshold, convert the packet loss rate data into the corresponding packet loss level and store it in the detection node packet loss level data table. From the detection node information table and the detection node packet loss level data table, perform data association according to the available zone dimension, perform frequency statistics on the packet loss level data of all detection nodes in each available zone, and obtain the distribution of the number of nodes at each packet loss level.

[0066] Get the preset network packet loss level weight coefficient and determine the network packet loss level for each detection node. According to the distribution of the network packet loss level of the detection node, use the weighted calculation method to obtain the network packet loss weighted score of each detection node. Summarize the network packet loss weighted scores of the detection nodes in the same cloud availability zone to obtain the comprehensive network packet loss score of the availability zone.

[0067] Exemplarily, in order to determine the weight coefficient of the network packet loss index data of each detection node, the hierarchical analysis method can be used. First, by means of expert scoring, the importance of indicators such as delay, jitter, and retransmission rate is compared in pairs to construct a judgment matrix. Then, the weight vector of each indicator is calculated using the matrix eigenvalue method, and a consistency check is performed until it passes. For example, the weight coefficients finally calculated are: delay 5, jitter 3, and retransmission rate 2. On this basis, the network packet loss index data of each detection node is weighted and fused to obtain a comprehensive score using a weighted average model. Then, the K-means clustering algorithm is used to perform cluster analysis on the comprehensive scores of all detection nodes. By calculating the Euclidean distance from each scoring sample to the cluster center and iteratively updating the cluster center until the clustering results converge, the detection nodes can eventually be divided into three packet loss levels: low, medium, and high. In each cloud availability zone, the frequency distribution of detection nodes with different packet loss levels is counted, and the frequency distribution is weighted and averaged according to the preset level weight coefficient (such as low 2, medium 3, high 5) to obtain the overall packet loss score of the network in the cloud availability zone, which is used for subsequent availability zone selection decisions. Through the above method, the multi-dimensional indicators are converted into a one-dimensional score, and the differences in the importance of different indicators and levels are taken into account, which can more comprehensively and reasonably evaluate the packet loss level of the cloud availability zone network.

[0068] S103, determining a network packet loss score threshold for each available zone network on the cloud according to the network area division, service type, and time period configuration parameters, and determining whether a network packet loss failure event has occurred in each available zone network on the cloud by comparing the network packet loss comprehensive score of each available zone network on the cloud with the network packet loss score threshold.

[0069] In this embodiment, in order to more accurately monitor the network packet loss of the cloud available zone network and promptly detect network packet loss failure events, a network packet loss failure monitoring system based on dynamic thresholds is designed and implemented. The system configures specific network packet loss score thresholds for each cloud available zone network according to parameters such as network area division, business type, and time period. By comparing the network packet loss comprehensive score with these thresholds in real time, the system can automatically determine and report network packet loss failure events.

[0070] Specifically, the network is divided into different areas according to the physical location, topology and other factors of the cloud availability zone network. Each area may have different network characteristics and business requirements. Identify and classify different types of services running on the cloud, such as Web services, database services, video streaming, etc. Different service types will have different sensitivities to network packet loss. Considering the periodic changes in network traffic, the time is divided into different time periods (such as peak hours, off-peak hours, night time, etc.). The network load and packet loss in different time periods will also be different. Based on historical data and business requirements, set a reasonable network packet loss score threshold for each cloud availability zone network, each service type and each time period. Deploy detection nodes in each cloud availability zone network to collect multi-dimensional network performance indicator data such as network packet loss, delay, and jitter in real time. Use a weighted fusion algorithm to process the collected data to obtain a comprehensive network packet loss score for each cloud availability zone network. The system compares the comprehensive network packet loss score of each cloud availability zone network with its corresponding threshold in real time. When the score exceeds the threshold, the system automatically determines it as a network packet loss failure event.

[0071] In this embodiment, by configuring a specific network packet loss score threshold for each cloud availability zone network and comparing the score with the threshold in real time, the system can more accurately detect network packet loss failure events and improve the fault detection rate. The dynamic threshold mechanism can automatically adjust the threshold according to changes in parameters such as network area division, business type and time period, thereby reducing the false alarm and missed alarm rate caused by improper fixed threshold settings.

[0072] In some embodiments, in the above step S103, determining the network packet loss score threshold of each available zone network on the cloud according to the network area division, service type and time period configuration parameters specifically includes:

[0073] In each available zone network on the cloud, a corresponding network area is divided for each user, and a first scoring coefficient of each network area is determined according to user feature information of the user corresponding to each network area;

[0074] Determine a second scoring coefficient for each service type according to the sensitivity of each service type to network packet loss;

[0075] Determine a third scoring coefficient for each service access time period according to the user traffic situation in the service access time period;

[0076] A weighted average algorithm is adopted to determine a network packet loss score threshold of each available zone network on the cloud through the first score coefficient, the second score coefficient and the third score coefficient.

[0077] In this embodiment, user information in each available zone network on the cloud is obtained, including user characteristics, network usage requirements and other data. Based on the obtained user characteristic information, a clustering algorithm is used to group users, and users with similar characteristics are divided into the same network area. The user characteristic information in each network area is analyzed to extract key characteristic parameters, such as bandwidth requirements, delay sensitivity, etc. Based on the key characteristic parameters of users in the network area and the availability of network resources, the first scoring coefficient of each network area is calculated.

[0078] The service type information and the corresponding network packet loss sensitivity data are obtained, and the sensitivity level of the service type to network packet loss is determined according to a preset sensitivity threshold, and the corresponding second scoring coefficient is determined for different sensitivity levels. Specifically, a machine learning algorithm is used to train the historical data of service type and network packet loss sensitivity to obtain a correlation model between service type and sensitivity, and the packet loss sensitivity of a new service type is dynamically predicted through the model, and its second scoring coefficient is determined.

[0079] Obtain business access log data and extract information such as user access time and access traffic. Divide the data into different time periods according to the access time, such as every hour, every day, etc. For each time period, count the total access traffic within the time period and calculate the proportion of traffic in each time period to the total traffic for the whole day. Use the traffic proportion of each time period as the third scoring coefficient of the time period and establish a mapping relationship between the time period and the scoring coefficient. In the subsequent business scoring, obtain the corresponding third scoring coefficient from the mapping relationship according to the time period of the current business access.

[0080] The first scoring coefficient, the second scoring coefficient, and the third scoring coefficient are used as input parameters, and the weighted average algorithm is used to calculate the network packet loss scoring threshold of each cloud availability zone network. If the network packet loss scoring threshold exceeds the preset warning value, it is judged that the network quality of the cloud availability zone network is at risk and network optimization and adjustment are required. According to the network packet loss scoring threshold, the quality of each cloud availability zone network is evaluated and sorted, and the availability zone network with a higher scoring threshold is selected as the preferred network resource to carry the service requests of key businesses and high-priority users.

[0081] Exemplarily, in order to determine the first scoring coefficient of each network area, a fuzzy comprehensive evaluation method can be used. First, a fuzzy evaluation matrix is ​​constructed based on the user's bandwidth demand, delay sensitivity, business importance and other characteristics. Then, the qualitative indicators are quantified into membership values ​​between 0 and 1 using the membership function. Next, the comprehensive membership is calculated by the weighted average method as the first scoring coefficient of the network area. For example, if the bandwidth demand membership of users in a certain area is 8, the delay sensitivity membership is 6, and the business importance membership is 9, then its first scoring coefficient is 8×3+6×3+9×4=78.

[0082] When determining the second scoring coefficient, the packet loss sensitivity of different service types can be scored based on expert experience, with scores between 0 and 10. Then, the scores of each service type are normalized, and the weight coefficient obtained is the second scoring coefficient. For example, the packet loss sensitivity scores of video service, game service and download service are 10, 8 and 6 respectively, and the normalized second scoring coefficients are 42, 33 and 25.

[0083] For the third scoring coefficient, we can statistically analyze the distribution of user traffic in different time periods and combine it with expert experience to assign a weight coefficient between 0 and 1 to each time period. For example, if a day is divided into four time periods: early morning, morning, afternoon, and evening, the corresponding weight coefficients are 1, 3, 3, and 3 respectively. Finally, we use the weighted average model and the above three scoring coefficients to calculate the packet loss score threshold of each cloud availability zone network.

[0084] Assuming the weights of the first, second, and third scoring coefficients are 4, 35, and 25 respectively, the packet loss scoring threshold = first scoring coefficient × 4 + second scoring coefficient × 35 + third scoring coefficient × 25. When the actual packet loss comprehensive score of the available zone network exceeds the threshold, it indicates that the network quality is poor and needs to be optimized.

[0085] S104, determining packet loss failure event data of a network in an available zone on the cloud where a network packet loss failure event has occurred, and determining a packet loss event level according to the packet loss failure event data.

[0086] In this embodiment, after a network packet loss failure event occurs in the cloud availability zone network, in order to more effectively assess the impact scope and severity of the failure, a packet loss failure event level determination system is designed and implemented. The system automatically determines the level of the packet loss event based on the packet loss failure event data that has occurred through a series of evaluation indicators and algorithms, so that corresponding countermeasures can be taken.

[0087] Specifically, when the system detects a network packet loss failure event in the cloud availability zone network, relevant failure data is collected immediately. This data may include the time and location (specific availability zone) of the failure, the affected business type, packet loss rate, duration, number of concurrent failures (such as increased latency, increased jitter), etc. Among them, the packet loss rate is used to measure the proportion of data packets lost during network transmission, the duration indicates the length of time from the occurrence to the end of the failure event, the affected business type is used to evaluate the different sensitivities of different business types to network packet loss, and the number of concurrent failures is used to evaluate whether other network failures have occurred in the same time period, and the correlation between these failures.

[0088] Based on the above evaluation indicators, an evaluation algorithm is designed and implemented to determine the level of packet loss failure events. The algorithm can use weighted summation, fuzzy evaluation, machine learning and other methods to convert the values ​​of each evaluation indicator into a comprehensive score, and then divide the failure events into different levels (such as severe, general, minor, etc.) according to the score. According to the output results of the evaluation algorithm, packet loss failure events are divided into different levels. Each level corresponds to different response measures and priorities. For example, a severe level of failure event may require the immediate activation of an emergency plan, notification of key users, and the deployment of resources for emergency repairs; while a minor level of failure event may only need to be recorded and monitored for subsequent optimization processing.

[0089] In this embodiment, by defining clear evaluation indicators and adopting scientific evaluation algorithms, the system can more accurately evaluate the level of network packet loss failure events in the cloud availability zone, reducing the errors and subjectivity of human judgment. According to the different levels of failure, the system can reasonably allocate resources for fault handling. For severe level failure events, more resources and manpower can be deployed for emergency repair; for minor level failure events, more flexible handling methods can be adopted.

[0090] In some embodiments, in the above step S104, determining the severity of packet loss in the cloud available zone network where the network packet loss failure event has occurred, and determining the level of the packet loss event according to the severity of the packet loss specifically includes:

[0091] Obtain historical packet loss data of the available zone network on the cloud, obtain a packet loss event indicator based on the historical packet loss data, and generate an original packet loss event log based on the packet loss event indicator;

[0092] According to the preset time window, the original packet loss event log is statistically analyzed to obtain the average packet loss rate and the maximum packet loss rate in each time window;

[0093] According to the average packet loss rate and maximum packet loss rate in all time windows, the severity of the network packet loss failure is determined to form a packet loss event training data set;

[0094] Inputting the packet loss event training data set into a decision tree model for training to obtain a packet loss event classification model;

[0095] Obtain packet loss failure event data of a network in an available zone on the cloud where a network packet loss failure event has occurred, input the packet loss failure event data into the packet loss event grading model for identification, and determine the packet loss event grade of the network in the available zone on the cloud where the network packet loss failure event has occurred.

[0096] In this embodiment, the historical packet loss data of the available area network on the cloud is obtained, and the packet loss data of different time periods are statistically analyzed to obtain indicators such as the average packet loss rate and the frequency of packet loss in each time period. According to the statistical analysis results of the historical packet loss data, the judgment threshold of the packet loss event is determined. If the packet loss rate or the frequency of packet loss in a certain time period exceeds the preset threshold, it is determined that a packet loss event has occurred in the time period. For the time period determined as a packet loss event, the packet loss data characteristics in the time period are further analyzed, and key indicators such as the start and end time of the packet loss, the peak value of the packet loss rate, and the packet loss event indicator data are extracted to generate the packet loss event indicator data. The packet loss event indicator data is classified by a clustering algorithm, and different types of packet loss events are identified according to the clustering results, such as packet loss caused by network congestion, packet loss caused by link failure, etc. For each type of packet loss event, based on its packet loss data characteristics and occurrence rules, a corresponding packet loss event detection model is constructed to monitor the network packet loss in real time and detect packet loss events in time. After the packet loss event is detected, the original packet loss event log in a standard format is automatically generated according to indicators such as the time of the event, the packet loss rate, etc., to record the detailed information of the event.

[0097] Get the preset time window parameters and determine the start and end time of the time window; according to the time window parameters, filter out the packet loss event records that fall within each time window from the original packet loss event log; for the packet loss event records in each time window, count the total number of packet loss events and the total number of data packets; calculate the average packet loss rate in each time window, the formula is: average packet loss rate = number of packet loss events in the window / total number of data packets in the window; calculate the maximum packet loss rate in each time window, and get it by finding the maximum value of the packet loss rate in the window. According to the preset packet loss fault severity threshold, determine whether the maximum packet loss rate exceeds the threshold. If it exceeds, it is considered that a serious packet loss fault has occurred. Use the support vector machine algorithm to train the packet loss fault severity classification model with the average packet loss rate and the maximum packet loss rate as features. Use the decision tree algorithm to train the packet loss event classification model with the average packet loss rate and the maximum packet loss rate as features to determine whether a packet loss event has occurred. Use the average packet loss rate, the maximum packet loss rate, the packet loss fault severity, and the packet loss event as sample attributes to construct a packet loss event training data set.

[0098] Based on the preprocessed training data set, the decision tree algorithm is used to build a packet loss event classification model. The hyperparameters of the decision tree model are optimized through the cross-validation method to improve the classification accuracy of the model. The trained decision tree model is used to predict and classify the newly occurring packet loss event data. If the confidence of the prediction result is lower than the preset threshold, the model retraining is triggered and the packet loss event classification model is updated.

[0099] The network monitoring system continuously obtains real-time packet loss data and related event information of the cloud availability zone network. When a packet loss failure is detected, the subsequent packet loss event classification process is triggered. For the acquired packet loss failure event data, key characteristic parameters such as the time of packet loss, duration, and packet loss rate are extracted and formatted into a standardized data form for input into the packet loss event classification model. The standardized packet loss event data is input into the packet loss event classification model to automatically identify and analyze the severity of the packet loss event and give the corresponding rating results.

[0100] Further, the step of determining the packet loss event level of the network in the cloud availability zone where the network packet loss failure event has occurred further includes:

[0101] Obtaining a network topology structure of the cloud availability zone network where the packet loss event level is a severe level, determining a connection relationship and network device information between host nodes of the cloud availability zone network according to the network topology structure, and forming a network topology map;

[0102] Based on the network topology diagram, a graph neural network model is used to determine a fault propagation characteristic pattern, and an optimal response measure to a network packet loss fault is determined according to the fault propagation characteristic pattern.

[0103] In this embodiment, the information of network devices with serious packet loss events in the cloud available zone network is obtained, and the network topology model is constructed by analyzing the connection relationship between network devices. According to the network topology model, the connectivity between host nodes is analyzed by graph theory algorithm to obtain the connection relationship matrix between host nodes. The configuration information and alarm log of network devices are analyzed by deep learning algorithm to determine whether there is an abnormality in the network device. If there is an abnormality, the abnormal device is marked in the topology map.

[0104] A graph neural network model is constructed based on the network topology graph, and the network nodes and links are mapped to the nodes and edges of the graph. The attributes of the nodes and edges include network attribute parameters such as network device type and link bandwidth. Based on historical network fault data, characteristic parameters such as fault propagation path, propagation time, and impact range are extracted, and the fault propagation characteristic pattern is obtained through training through the graph neural network model. When a packet loss failure occurs in the network, the current network topology structure and fault alarm information are obtained and input into the trained graph neural network model to predict the fault propagation path and impact range. According to the predicted fault propagation characteristic pattern, the fault knowledge base is searched for matches to obtain candidate fault corresponding measures. The candidate measures include Link load balancing, dynamic routing, etc.; use reinforcement learning algorithms to evaluate candidate fault response measures, with evaluation indicators including packet loss rate recovery time, network throughput, etc., and obtain the optimal measure with the highest evaluation benefit through multiple iterations; send the optimal fault response measure to the network controller, and dynamically adjust the network forwarding rules and link strategies through software-defined network technology to achieve rapid location and recovery of network packet loss faults; continuously monitor the network status, obtain network performance indicators after fault recovery, verify the effectiveness of measures through A / B testing, etc., optimize the graph neural network model and fault knowledge base according to the feedback results, and continuously improve the intelligence level of fault diagnosis and recovery.

[0105] S105, based on the packet loss event level and the attribute information of each operation and maintenance personnel, determine different network fault briefing generation rules and different network fault briefing push rules for each operation and maintenance personnel, and generate and push corresponding network fault briefings for each operation and maintenance personnel according to the different network fault briefing generation rules and different network fault briefing push rules.

[0106] In this embodiment, in order to improve the response efficiency and processing ability of the operation and maintenance team to network fault events, a personalized network fault briefing generation and push system is designed and implemented. The system customizes different network fault briefing generation rules and push rules for each operation and maintenance personnel based on the determined packet loss event level and the attribute information of each operation and maintenance personnel (such as skills, scope of responsibilities, preference settings, etc.). By generating and pushing personalized network fault briefings, the system can ensure that the operation and maintenance personnel obtain fault information related to their responsibilities and interests in a timely manner, so as to make responses and decisions more quickly.

[0107] Specifically, the attribute information of the operation and maintenance personnel is collected, including but not limited to: skill expertise, the knowledge and experience of the operation and maintenance personnel in specific technical or business fields; scope of responsibilities, the specific network areas, business types or system components that the operation and maintenance personnel are responsible for; preference settings, the personal preferences of the operation and maintenance personnel for briefing content, format, push method, etc.

[0108] Based on the packet loss event level and the attribute information of the operation and maintenance personnel, the system formulates personalized briefing generation rules for each operation and maintenance personnel. These rules include: content screening, filtering out relevant fault information based on the operation and maintenance personnel's scope of responsibilities and skill expertise; priority sorting, for multiple fault events, sorting them according to the event level and the urgency of the operation and maintenance personnel's responsibilities; formatting customization: customize the format, charts, colors, etc. of the briefing according to the operation and maintenance personnel's preferences.

[0109] The system also formulates personalized briefing push rules for each operation and maintenance personnel. These rules include: push time, selecting the appropriate push time according to the operation and maintenance personnel's working hours and preferences; push method, pushing briefings through various methods such as email, SMS, instant messaging, mobile applications, etc.; push frequency, setting the appropriate push frequency according to the urgency of the fault event and the needs of the operation and maintenance personnel.

[0110] According to the established personalized briefing generation rules and push rules, the system generates a corresponding network fault briefing for each operation and maintenance personnel, and pushes it through the selected method.

[0111] In this embodiment, by providing maintenance personnel with fault information that is closely related to their duties and interests, the system can significantly improve the efficiency of the maintenance team's response to network fault events. Maintenance personnel can more quickly identify and handle fault events related to themselves, reducing unnecessary interference and delays. Personalized briefings not only contain basic information about fault events, but can also include analysis and suggestions for maintenance personnel's skills and expertise, which helps maintenance personnel understand the fault situation more comprehensively and make more accurate decisions.

[0112] In some embodiments, in the above step S105, based on the packet loss event level and the attribute information of each operation and maintenance personnel, different network fault report generation rules and different network fault report push rules are determined for each operation and maintenance personnel, specifically including:

[0113] Acquire attribute information of each operation and maintenance personnel in a pre-established operation and maintenance personnel information database, wherein the attribute information includes level, professional scope and responsibility attributes;

[0114] A K-means algorithm is used to perform cluster analysis on the attribute information, and operation and maintenance personnel with similar attribute information are classified into the same category, and a corresponding network fault briefing template is set for each category;

[0115] The level, professional scope, and responsibilities of the operation and maintenance personnel and the historical packet loss event data are used as the input features of the decision tree. The personalized network fault briefing generation rules and network fault briefing push rules are used as the output of the decision tree. By training the decision tree model, a network fault briefing rule setting model is formed.

[0116] According to the network fault briefing rule setting model and the network fault briefing template, different network fault briefing generation rules and different network fault briefing push rules are determined for each operation and maintenance personnel through the packet loss event level and the attribute information of each operation and maintenance personnel.

[0117] In this embodiment, a pre-established operation and maintenance personnel information database is obtained, and the level, professional scope and responsibility attribute information of each operation and maintenance personnel are extracted from the database; the extracted attribute information is converted into a numerical vector, and an operation and maintenance personnel attribute feature matrix is ​​constructed; the K-means algorithm is used to perform cluster analysis on the operation and maintenance personnel attribute feature matrix, and the operation and maintenance personnel with similar attributes are divided into the same category by calculating the distance between each data point and the cluster center; for each operation and maintenance personnel category obtained by clustering, the common attribute characteristics of the operation and maintenance personnel in this category are analyzed, and a corresponding network fault briefing template is set according to the attribute characteristics; when a network fault occurs, the corresponding operation and maintenance personnel category is matched according to the fault type and the profession to which the faulty equipment belongs; the network fault briefing template corresponding to the category is obtained, the fault information is automatically filled in, and a targeted fault briefing is generated.

[0118] According to the level, professional scope and job attributes of the operation and maintenance personnel, the corresponding weight coefficient is obtained; based on the historical packet loss event data, the efficiency and accuracy of different operation and maintenance personnel in handling faults are counted; the operation and maintenance personnel attributes and historical data are used as the input features of the decision tree, and the personalized briefing rules are used as the output; the decision tree model is constructed using the ID3 algorithm, and the optimal partitioning attribute is selected through information gain; at each decision node, the data set is divided into multiple subsets according to the attribute value; sub-decision trees are recursively constructed until all samples belong to the same category or cannot be further divided; according to the path of the decision tree, personalized network fault briefing generation rules and push rules are generated.

[0119] According to the network fault briefing rule setting model and network fault briefing template, combined with the current packet loss event level and the event-related operation and maintenance personnel attribute information, personalized network fault briefing content is dynamically generated. According to the push rules and operation and maintenance personnel attributes in the network fault briefing rule setting model, determine which operation and maintenance personnel the briefing content should be pushed to, and form a personalized network fault briefing push list. Send the personalized network fault briefing content to the corresponding operation and maintenance personnel according to the push list, and dynamically optimize the network fault briefing rule setting model based on feedback to continuously improve the accuracy and timeliness of network fault briefings.

[0120] Exemplarily, in order to obtain the attribute information of the operation and maintenance personnel, a database containing the basic information of all operation and maintenance personnel can be established in advance. The records of each operation and maintenance personnel in the database include: level (such as junior, intermediate, senior), professional scope (such as network, system, security, etc.) and responsibility attributes (such as duty, inspection, emergency, etc.). Then, the K-means clustering algorithm is used to analyze these attribute information. The algorithm first randomly selects k cluster centers, and then iteratively assigns each data point to the nearest cluster center, and updates the position of the cluster center until the cluster center no longer changes. Assume that the operation and maintenance personnel are divided into 5 categories, and the attribute characteristics of each category of personnel are similar, such as the level is intermediate and the professional scope is network. For each category, a different network fault briefing template is set, and the content and format in the template are customized according to the characteristics of the personnel in that category. Next, the C5 algorithm is used to train the decision tree model, taking the attributes of the operation and maintenance personnel and the historical packet loss event data as features, and using personalized briefing generation rules (such as detailed briefing content for junior network operation and maintenance personnel, and highlighting the impact of the fault for senior security operation and maintenance personnel) and briefing push rules (such as push within 5 minutes for serious faults and within 1 hour for general faults) as the output of the decision tree. When a new packet loss fault occurs, the decision tree model determines the briefing generation rules and push rules corresponding to each person based on the severity of the fault and the attributes of the operation and maintenance personnel, and then combines the preset briefing template to automatically generate personalized network fault briefings, and send them to the relevant operation and maintenance personnel in a timely manner according to the push rules, so that the fault can be quickly handled.

[0121] As an application example of the above embodiment, Figure 2 As shown in the figure, the network monitoring system (such as the open source zabbix system, etc.) deploys monitoring agents in the available zones to collect network packet loss data between the available zones in real time through a simple ping full interconnection method; when the network packet loss in the available zone on the cloud reaches the set threshold, the network monitoring system will trigger an alarm; the network monitoring system calls back the push platform interface; the push platform pulls the monitoring data of the network monitoring system through the interface; by pulling the monitoring data, combined with the preset threshold and judgment conditions of the push platform, the network fault conclusion is output through calculation and comparison, and written to the database. Outputting the network fault conclusion will trigger the push module, generate a network fault briefing and push it. Before pushing the network fault briefing, first find the database to pull the previous fault conclusion and push time for comparison. Within 5 minutes, the same conclusion will not be pushed repeatedly. Call the third-party webhook interface to complete the push of the network fault briefing and notify the on-duty personnel by phone.

[0122] Reference Figure 3 An embodiment of the present invention provides a system 3 for processing network packet loss in an available zone on a cloud, and the system 3 specifically includes:

[0123] The first processing module 301 is used to deploy a plurality of detection nodes in the available zone networks on each cloud, and obtain multi-dimensional network packet loss indicator data of the available zone networks on each cloud according to the detection nodes;

[0124] The second processing module 302 is used to use a weighted fusion algorithm to process the multi-dimensional network packet loss indicator data of all detection nodes of each cloud available zone network, and obtain a comprehensive network packet loss score of each cloud available zone network;

[0125] The third processing module 303 is used to determine the network packet loss score threshold of each cloud available zone network according to the network area division, service type and time period configuration parameters, and determine whether a network packet loss failure event has occurred in each cloud available zone network by comparing the network packet loss comprehensive score of each cloud available zone network with the network packet loss score threshold;

[0126] The fourth processing module 304 is used to determine the packet loss failure event data of the cloud available zone network where the network packet loss failure event has occurred, and determine the packet loss event level according to the packet loss failure event data;

[0127] The fifth processing module 305 is used to determine different network fault briefing generation rules and different network fault briefing push rules for each operation and maintenance personnel based on the packet loss event level and the attribute information of each operation and maintenance personnel, and generate and push corresponding network fault briefings for each operation and maintenance personnel according to different network fault briefing generation rules and different network fault briefing push rules.

[0128] It is understandable that if Figure 1 The contents of the embodiment of the method for processing network packet loss in the cloud available zone shown in the figure are applicable to the embodiment of the system for processing network packet loss in the cloud available zone. The functions specifically implemented by the embodiment of the system for processing network packet loss in the cloud available zone are similar to those in the embodiment of the method for processing network packet loss in the cloud available zone shown in the figure. Figure 1 The method for processing packet loss in the cloud availability zone network shown in the embodiment is the same as that in the embodiment shown in the embodiment, and the beneficial effect achieved is the same as that in the embodiment shown in the embodiment. Figure 1 The beneficial effects achieved by the embodiment of the method for processing network packet loss in the cloud availability zone shown are also the same.

[0129] It should be noted that the information interaction, execution process and other contents between the above-mentioned systems are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0130] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0131] Reference Figure 4 The embodiment of the present invention further provides a computer device 4, comprising: a memory 402 and a processor 401 and a computer program 403 stored on the memory 402. When the computer program 403 is executed on the processor 401, a method for processing network packet loss in an available zone on the cloud as described in any one of the above methods is implemented.

[0132] The computer device 4 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 4 may include, but is not limited to, a processor 401 and a memory 402. Those skilled in the art will appreciate that Figure 4 It is only an example of the computer device 4 and does not constitute a limitation on the computer device 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0133] The processor 401 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0134] In some embodiments, the memory 402 may be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 402 may also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 4. Further, the memory 402 may also include both an internal storage unit and an external storage device of the computer device 4. The memory 402 is used to store an operating system, an application program, a boot loader (Boot Loader), data, and other programs, such as the program code of the computer program, etc. The memory 402 may also be used to temporarily store data that has been output or is to be output.

[0135] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for processing network packet loss in an available zone on the cloud as described in any one of the above methods is implemented.

[0136] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0137] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0138] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0139] In the embodiments disclosed in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0140] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Claims

1. A method for processing packet loss in a cloud availability zone network, characterized in that: The method specifically comprises: Deploy several detection nodes in each available zone network on each cloud, and obtain multi-dimensional network packet loss indicator data of each available zone network on each cloud according to the detection nodes; A weighted fusion algorithm is used to process the multi-dimensional network packet loss indicator data of all detection nodes in each cloud availability zone network to obtain a comprehensive network packet loss score for each cloud availability zone network. Determine the network packet loss score threshold of each cloud availability zone network based on network area division, service type, and time period configuration parameters. Determine whether a network packet loss failure event has occurred in each cloud availability zone network by comparing the network packet loss comprehensive score of each cloud availability zone network with the network packet loss score threshold. Determine packet loss failure event data of a network in an available zone on the cloud where a network packet loss failure event has occurred, and determine a packet loss event level according to the packet loss failure event data; Based on the packet loss event level and the attribute information of each operation and maintenance personnel, different network fault briefing generation rules and different network fault briefing push rules are determined for each operation and maintenance personnel, and corresponding network fault briefings are generated and pushed for each operation and maintenance personnel according to the different network fault briefing generation rules and different network fault briefing push rules.

2. The method according to claim 1, characterized in that The obtaining of network performance data of each available zone network on the cloud according to the detection node also includes: Obtain the topology information and path dynamic change information of each available zone network on the cloud, and determine the initial detection path and detection frequency parameters according to the topology information and the path dynamic change information; The ARIMA time series analysis algorithm is used to model and analyze the historical detection data of the network in each available zone on the cloud to obtain the probability distribution of packet loss under different paths and frequencies. According to the initial detection path, the detection frequency parameters and the packet loss probability distribution under different paths and at different frequencies, an integer programming detection strategy optimization model is constructed with the minimization of the overall packet loss monitoring blind area as the optimization goal. The integer programming detection strategy optimization model is used to determine the optimal detection strategy for each available zone network on the cloud.

3. The method according to claim 1, characterized in that The weighted fusion algorithm is used to process the multi-dimensional network packet loss index data of all detection nodes in each cloud available zone network, and obtain the network packet loss comprehensive score of each cloud available zone network, specifically including: The hierarchical analysis method is used to determine the weight coefficient of the network packet loss index data of each detection node; According to the preset weighted fusion algorithm model and the weight coefficient, the network packet loss index data of each detection node is weightedly calculated to obtain a comprehensive network packet loss score for each detection node; The K-means clustering algorithm is used to perform cluster analysis on the comprehensive scores of network packet loss of all detection nodes to determine the network packet loss level data of each detection node; Obtain network packet loss level data of all detection nodes of each available zone network on each cloud, and obtain the distribution of network packet loss levels of detection nodes of each available zone network on each cloud by performing frequency statistics on the network packet loss level data; According to the preset network packet loss level weight coefficient, the distribution of the network packet loss level of the detection node is weightedly calculated to obtain a comprehensive network packet loss score for each available zone network on the cloud.

4. The method according to claim 1, characterized in that The method of determining the network packet loss score threshold of each available zone network on the cloud according to the network area division, service type and time period configuration parameters specifically includes: In each available zone network on the cloud, a corresponding network area is divided for each user, and a first scoring coefficient of each network area is determined according to user feature information of the user corresponding to each network area; Determine a second scoring coefficient for each service type according to the sensitivity of each service type to network packet loss; Determine a third scoring coefficient for each service access time period according to the user traffic situation in the service access time period; A weighted average algorithm is adopted to determine a network packet loss score threshold of each available zone network on the cloud through the first score coefficient, the second score coefficient and the third score coefficient.

5. The method according to claim 1, characterized in that The determining of the severity of packet loss in the cloud available zone network where the network packet loss failure event has occurred, and determining the level of the packet loss event according to the severity of the packet loss specifically includes: Obtain historical packet loss data of the available zone network on the cloud, obtain a packet loss event indicator based on the historical packet loss data, and generate an original packet loss event log based on the packet loss event indicator; According to the preset time window, the original packet loss event log is statistically analyzed to obtain the average packet loss rate and the maximum packet loss rate in each time window; According to the average packet loss rate and maximum packet loss rate in all time windows, the severity of the network packet loss failure is determined to form a packet loss event training data set; Inputting the packet loss event training data set into a decision tree model for training to obtain a packet loss event classification model; Obtain packet loss failure event data of a network in an available zone on the cloud where a network packet loss failure event has occurred, input the packet loss failure event data into the packet loss event grading model for identification, and determine the packet loss event grade of the network in the available zone on the cloud where the network packet loss failure event has occurred.

6. The method according to claim 5, characterized in that The step of determining the packet loss event level of the cloud available zone network where the network packet loss failure event has occurred further includes: Obtaining a network topology structure of the cloud availability zone network where the packet loss event level is a severe level, determining a connection relationship and network device information between host nodes of the cloud availability zone network according to the network topology structure, and forming a network topology map; Based on the network topology diagram, a graph neural network model is used to determine a fault propagation characteristic pattern, and an optimal response measure to a network packet loss fault is determined according to the fault propagation characteristic pattern.

7. The method according to any one of claims 1 to 6, characterized in that Based on the packet loss event level and the attribute information of each operation and maintenance personnel, different network fault briefing generation rules and different network fault briefing push rules are determined for each operation and maintenance personnel, specifically including: Acquire attribute information of each operation and maintenance personnel in a pre-established operation and maintenance personnel information database, wherein the attribute information includes level, professional scope and responsibility attributes; A K-means algorithm is used to perform cluster analysis on the attribute information, and operation and maintenance personnel with similar attribute information are classified into the same category, and a corresponding network fault briefing template is set for each category; The level, professional scope, and responsibilities of the operation and maintenance personnel and the historical packet loss event data are used as the input features of the decision tree. The personalized network fault briefing generation rules and network fault briefing push rules are used as the output of the decision tree. By training the decision tree model, a network fault briefing rule setting model is formed. According to the network fault briefing rule setting model and the network fault briefing template, different network fault briefing generation rules and different network fault briefing push rules are determined for each operation and maintenance personnel through the packet loss event level and the attribute information of each operation and maintenance personnel.

8. A system for processing network packet loss in a cloud availability zone, characterized in that: The system specifically comprises: A first processing module is used to deploy a plurality of detection nodes in each available zone network on each cloud, and obtain multi-dimensional network packet loss indicator data of each available zone network on each cloud according to the detection nodes; The second processing module is used to use a weighted fusion algorithm to process the multi-dimensional network packet loss indicator data of all detection nodes of each cloud available zone network, and obtain a comprehensive network packet loss score for each cloud available zone network; The third processing module is used to determine the network packet loss score threshold of each cloud availability zone network according to the network area division, service type and time period configuration parameters, and determine whether a network packet loss failure event has occurred in each cloud availability zone network by comparing the network packet loss comprehensive score of each cloud availability zone network with the network packet loss score threshold; A fourth processing module is used to determine packet loss failure event data of the cloud available zone network where the network packet loss failure event has occurred, and determine the packet loss event level according to the packet loss failure event data; The fifth processing module is used to determine different network fault briefing generation rules and different network fault briefing push rules for each operation and maintenance personnel based on the packet loss event level and the attribute information of each operation and maintenance personnel, and generate and push corresponding network fault briefings for each operation and maintenance personnel according to different network fault briefing generation rules and different network fault briefing push rules.

9. A computer device, characterized in that: include: A memory and a processor and a computer program stored in the memory, when the computer program is executed on the processor, implements the method for processing network packet loss in a cloud availability zone as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the method for processing network packet loss in an available area on the cloud as described in any one of claims 1 to 7 is implemented.

Images