CN118694803A - Cloud platform system traffic dynamic balancing processing method and device - Google Patents

Abstract

The application provides a method and a device for dynamically balancing flow of a cloud platform system, wherein the method comprises the following steps: monitoring real-time flow data in the cloud platform system, performing time sequence model analysis on the real-time flow data in a set time period, and determining a corresponding flow change trend; performing cluster analysis on a user access mode and a user access service corresponding to the real-time flow data, determining access behavior characteristics of the user, and determining a corresponding access behavior change trend according to the access behavior characteristics and through a preset association rule mining algorithm; inputting the flow change trend and the access behavior change trend into a set decision tree regression prediction model, and dynamically adjusting the resource allocation of the cloud platform system when a flow peak or an abnormal event is detected; the application can effectively improve the flexibility and response speed of the system, optimize the resource utilization rate and ensure the stable operation of the system under the peak flow.

Description

Cloud platform system traffic dynamic balancing processing method and device

Technical Field

The present application relates to the field of data processing, and specifically to a method and device for dynamically balancing traffic in a cloud platform system.

Background Art

In the Internet age, system traffic has become a difficult indicator to predict. Many Internet companies will predict the traffic of a certain activity in advance and manually add machines in advance to meet the peak traffic demand. However, traffic estimates often exceed expectations, and this method cannot completely solve the problem.

Current traffic management mainly relies on manual judgment and static configuration. This approach has many limitations. First, manual judgment of traffic peaks often relies on experience and historical data, but changes in Internet traffic are highly uncertain and instantaneous, and traditional methods are difficult to accurately predict. Second, statically configured resources are often stretched when faced with sudden traffic peaks, resulting in system performance degradation or even crashes.

In terms of traffic identification, existing systems mostly use preset rules and fixed thresholds for traffic monitoring and early warning. This method has a slow response speed when facing rapidly changing traffic and cannot adjust resource allocation in time. At the same time, the accuracy of traffic identification is also limited, and it is impossible to accurately distinguish between normal and abnormal traffic, which affects the efficiency and accuracy of resource scheduling.

In terms of resource expansion, traditional systems usually require manual intervention to add machines or configure resources in advance. This not only increases the cost and complexity of operation and maintenance, but also has the problem of low resource utilization. Since resource configuration cannot be adjusted dynamically in real time, the system often has a large number of idle resources during non-peak hours, causing waste. When traffic surges, the lag in resource expansion will lead to system performance bottlenecks and fail to meet user needs.

In general, the existing traffic management and resource expansion methods have many shortcomings and are difficult to cope with the ever-changing traffic demands in the Internet era. By introducing methods for dynamically identifying traffic and dynamically expanding services, the flexibility and response speed of the system can be effectively improved, resource utilization can be optimized, and user experience can be improved. However, this also puts higher requirements on technology and system architecture, requiring comprehensive improvements in data analysis, automated scheduling, and system stability.

Summary of the invention

In response to the problems in the prior art, the present application provides a method and device for dynamically balancing the traffic of a cloud platform system, which can effectively improve the flexibility and response speed of the system, optimize resource utilization, and ensure the stable operation of the system under peak traffic.

In order to solve at least one of the above problems, the present application provides the following technical solutions:

In a first aspect, the present application provides a method for dynamically balancing traffic in a cloud platform system, comprising:

Monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend;

Performing cluster analysis on user access patterns and user access services corresponding to the real-time traffic data, determining user access behavior characteristics based on the results of the cluster analysis, and determining corresponding access behavior change trends based on the access behavior characteristics and the correlation between user access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm;

The traffic change trend and the access behavior change trend are input into a set decision tree regression prediction model, and whether there is a traffic peak or abnormal event is determined according to the output of the decision tree regression prediction model. When the traffic peak or abnormal event is detected, the resource allocation of the cloud platform system is dynamically adjusted according to the traffic change demand output by the decision tree regression prediction model, and after the real-time traffic data of the cloud platform system drops back to a normal level, excess resources are released until the standard configuration is restored.

Furthermore, the real-time traffic data in the monitoring cloud platform system is analyzed by a time series model for the real-time traffic data within a set time period to determine the corresponding traffic change trend, including:

Collect user access traffic data from the cloud platform system in real time, clean and convert the user access traffic data, and convert the user access traffic data into a time series format;

The optimal parameter combination with the smallest prediction error of the ARIMA time series model on the user access flow data converted into a time series format is determined through grid search, and the flow change trend is determined according to the output of the ARIMA time series model.

Furthermore, performing cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data, and determining the user access behavior characteristics according to the results of the cluster analysis, includes:

Acquire user access patterns and user access services corresponding to the real-time traffic data;

According to the user access mode, the user access service and the preset K-Means clustering algorithm, the users are divided into different types of groups, and the user access behavior characteristics are determined, wherein each group has a unique access behavior characteristic.

Further, the determining of the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the user's access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm includes:

Determine the access behavior of different user groups based on user access behavior characteristics analysis;

The Apriori association rule mining algorithm is used to determine the correlation between the access behaviors of different user groups in their respective business scenarios on the user access data set, and the corresponding access behavior change trend is determined based on the correlation.

Furthermore, the flow change trend and the access behavior change trend are input into a decision tree regression prediction model, and judging whether there is a flow peak or abnormal event according to the output of the decision tree regression prediction model, includes:

Training a decision tree regression prediction model according to historical sample data until an optimal prediction rule is formed, and inputting the traffic change trend and the access behavior change trend into a set decision tree regression prediction model;

The presence of a traffic peak or abnormal event is determined based on the comparison between the predicted probability value output by the decision tree regression prediction model and the value of a preset threshold.

Furthermore, when the traffic peak or abnormal event is detected, the resource allocation of the cloud platform system is dynamically adjusted according to the traffic change demand output by the decision tree regression prediction model, including:

When a traffic peak or an abnormal event is detected, the key factors causing the abnormality are determined according to the output of the decision tree regression prediction model;

The resource allocation strategy of the cloud platform system is dynamically adjusted according to the traffic change demand predicted by the decision tree regression model and the key factors causing the anomaly.

Furthermore, after the real-time traffic data of the cloud platform system falls back to a normal level, the redundant resources are released until the standard configuration is restored, including:

Monitoring real-time traffic data of the cloud platform system;

After monitoring the real-time traffic data of the cloud platform system and returning to normal levels, the number of computing nodes and storage space are gradually reduced until the system's resource configuration is fully restored to a pre-set standard state.

In a second aspect, the present application provides a cloud platform system flow dynamic balancing processing device, comprising:

A traffic change determination module is used to monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend;

An access behavior determination module, used to perform cluster analysis on user access patterns and user access services corresponding to the real-time traffic data, determine the user's access behavior characteristics according to the results of the cluster analysis, and determine the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the user's access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm;

An abnormal prediction module is used to input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and determine whether there is a traffic peak or abnormal event based on the output of the decision tree regression prediction model. When the traffic peak or abnormal event is detected, the resource allocation of the cloud platform system is dynamically adjusted according to the traffic change demand output by the decision tree regression prediction model, and after the real-time traffic data of the cloud platform system drops back to a normal level, excess resources are released until the standard configuration is restored.

In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the cloud platform system traffic dynamic balancing processing method are implemented.

In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the cloud platform system traffic dynamic balancing processing method.

In a fifth aspect, the present application provides a computer program product, including a computer program/instruction, which, when executed by a processor, implements the steps of the cloud platform system traffic dynamic balancing processing method.

It can be seen from the above technical scheme that the present application provides a method and device for dynamically balancing the flow of a cloud platform system. By monitoring the real-time flow data in the cloud platform system, a time series model analysis is performed on the real-time flow data within a set time period to determine the corresponding flow change trend; a cluster analysis is performed on the user access patterns and user access services corresponding to the real-time flow data to determine the user's access behavior characteristics, and the corresponding access behavior change trend is determined based on the access behavior characteristics and a preset association rule mining algorithm; the flow change trend and the access behavior change trend are input into a set decision tree regression prediction model, and when a flow peak or abnormal event is detected, the resource allocation of the cloud platform system is dynamically adjusted, thereby effectively improving the flexibility and response speed of the system, optimizing resource utilization, and ensuring the stable operation of the system under peak flow.

BRIEF DESCRIPTION OF THE DRAWINGS

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

FIG1 is a flow chart of a method for dynamically balancing flow in a cloud platform system according to an embodiment of the present application;

FIG2 is a second flow chart of a method for dynamically balancing flow of a cloud platform system in an embodiment of the present application;

FIG3 is a third flow chart of a method for dynamically balancing flow of a cloud platform system in an embodiment of the present application;

FIG4 is a fourth flow chart of a method for dynamically balancing flow of a cloud platform system in an embodiment of the present application;

FIG5 is a fifth flow chart of a method for dynamically balancing flow of a cloud platform system in an embodiment of the present application;

FIG6 is a sixth flow chart of a method for dynamically balancing flow of a cloud platform system in an embodiment of the present application;

FIG. 7 is a seventh flow chart of a method for dynamically balancing flow of a cloud platform system in an embodiment of the present application;

FIG8 is a structural diagram of a cloud platform system flow dynamic balancing processing device in an embodiment of the present application;

FIG. 9 is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

Reference numerals:

Electronic device 9600, central processing unit 9100, communication module 9110, antenna 9111, input unit 9120, audio processor 9130, speaker 9131, microphone 9132, memory 9140, buffer memory 9141, application/function storage unit 9142, data storage unit 9143, driver program storage unit 9144, display 9160, power supply 9170.

DETAILED DESCRIPTION

In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

Taking into account the many deficiencies in the traffic management and resource expansion methods in the prior art, which are difficult to cope with the problem of rapidly changing traffic demands in the Internet era, the present application provides a method and device for dynamically balancing traffic in a cloud platform system. By monitoring the real-time traffic data in the cloud platform system, a time series model analysis is performed on the real-time traffic data within a set time period to determine the corresponding traffic change trend; a cluster analysis is performed on the user access patterns and user access services corresponding to the real-time traffic data to determine the user's access behavior characteristics, and the corresponding access behavior change trend is determined based on the access behavior characteristics and a preset association rule mining algorithm; the traffic change trend and the access behavior change trend are input into a set decision tree regression prediction model, and when a traffic peak or abnormal event is detected, the resource allocation of the cloud platform system is dynamically adjusted, thereby effectively improving the flexibility and response speed of the system, optimizing resource utilization, and ensuring the stable operation of the system under peak traffic.

In order to effectively improve the flexibility and response speed of the system, optimize resource utilization, and ensure the stable operation of the system under peak traffic, the present application provides an embodiment of a method for dynamically balancing traffic on a cloud platform system. Referring to FIG. 1 , the method for dynamically balancing traffic on a cloud platform system specifically includes the following contents:

Step S101: monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend;

Optionally, in this embodiment, step S101 involves monitoring of real-time traffic data and time series model analysis to determine traffic change trends. The technical process of this step involves data collection, preprocessing, model building and prediction, etc., aiming to improve the predictability of traffic management and the optimization capability of resource allocation.

First, in step S101, the system implements monitoring of real-time traffic data in the cloud platform. Real-time traffic data includes the number of user requests, data transmission volume, network bandwidth usage, etc. These data are collected and stored in real time through cloud platform monitoring tools (such as Prometheus and Grafana). The system configures a monitoring agent to continuously capture traffic information of each server node and aggregate the data into a central database for subsequent analysis.

Next, the system preprocesses the real-time traffic data within the set time period. This embodiment includes operations such as data cleaning, denoising, missing value filling, and data normalization to ensure data quality and consistency. For example, for incomplete traffic data records, the system can use interpolation or moving average methods to fill missing values, and use standardization methods to scale the data to a uniform range to eliminate the impact of magnitude differences.

After data preprocessing is completed, the system performs time series model analysis on the preprocessed traffic data. Time series analysis is a statistical method used to analyze and predict trends and cyclical fluctuations in time series data. The system first performs a visual analysis of the time series data to identify the basic characteristics of the data, such as trends, seasonality, and random fluctuations. Next, the system selects an appropriate time series prediction model, such as ARIMA (autoregressive integrated moving average model), SARIMA (seasonal autoregressive integrated moving average model), or LSTM (long short-term memory network), and builds and trains the model based on the data characteristics. For example, for traffic data with obvious seasonal fluctuations, the system can select the SARIMA model to capture the cyclical changes in the data by setting seasonal parameters.

After the model is built, the system uses the trained time series model to predict the traffic data within the set time period and determine the corresponding traffic change trend. The system inputs historical traffic data, and the model generates traffic forecasts for a period of time in the future based on the historical data pattern. For example, the system can predict the hourly traffic changes in the next week and generate a traffic trend chart. In this way, the system can identify traffic peaks and troughs in advance, thereby providing data support for resource scheduling and optimization.

This embodiment solves the limitation of traditional traffic management methods in dealing with dynamic changes and complex trends. Through real-time monitoring and time series analysis, the system can timely capture subtle fluctuations in traffic changes and make accurate predictions based on historical data, providing a scientific basis for resource management and optimization of the cloud platform. For example, when predicting future traffic peaks, the system can expand resources in advance to ensure service quality and user experience; during traffic troughs, the system can perform resource recovery and energy-saving operations to reduce operating costs.

The technical effect of this embodiment is significant. Through time series model analysis, the system can maintain efficient and accurate prediction capabilities in traffic forecasting and trend identification. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and predict traffic changes, and maintain high accuracy even in the face of complex and changeable traffic patterns. For example, when the system analyzes real-time traffic data, no matter how complex the traffic changes, the system can accurately predict future traffic trends through time series models, thereby adjusting resource allocation in a timely manner and improving the stability and performance of the overall system.

In general, step S101 solves the problems of dynamic changes and complex trends in traffic management through real-time traffic data monitoring and time series model analysis. Through the collaborative work of data collection, preprocessing, model building and prediction, the system not only improves the accuracy of traffic prediction, but also ensures the optimization and efficiency of resource scheduling. In the traffic management practice of the cloud platform, this technical solution has significantly improved the efficiency of prediction and decision-making, effectively reduced operational risks, and ensured the stable operation of the system. In actual applications, the cloud platform can more comprehensively identify and respond to traffic changes through this solution, improve the overall resource management level, and ensure service quality and user satisfaction.

Step S102: performing cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data, determining the user's access behavior characteristics according to the results of the cluster analysis, and determining the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the user's access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm;

Optionally, in this embodiment, step S102 involves clustering analysis of user access patterns and user access services corresponding to real-time traffic data to determine the user's access behavior characteristics, and analyzing the correlation between user behaviors through an association rule mining algorithm to determine the access behavior change trend. The technical process of this step includes cluster analysis, feature extraction, and association rule mining, aiming to improve the understanding and prediction capabilities of user behavior.

First, in step S102, the system performs cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data. Cluster analysis is an unsupervised learning method used to group similar objects in order to better understand the structure and characteristics of the data. The system first extracts user access records from the real-time traffic data, including information such as access time, access frequency, access pages, and service types. Then, the system uses algorithms such as K-Means, DBSCAN, or hierarchical clustering to perform cluster analysis on the user access data. For example, the system can use the K-Means algorithm to divide users into different groups (such as high-frequency access users, low-frequency access users, high-frequency users of specific services, etc.) according to their access frequency and service type to identify the access patterns of different types of users.

After the cluster analysis is completed, the system determines the user's access behavior characteristics based on the clustering results. Access behavior characteristics include the user's access frequency, access time preference, service type preference, etc. These characteristics can intuitively reflect the behavioral characteristics of users in different groups through statistical analysis and data visualization of clustering results. For example, the system can find that the frequency of access to video streaming services by users in a certain group in the evening is significantly higher than that in other periods, thereby determining the access time preference and service type preference of users in this group.

Next, the system uses a preset association rule mining algorithm to determine the correlation between users' access behaviors in different business scenarios on the user access dataset. Association rule mining is a data mining technique used to discover frequent patterns and associations between different items in a dataset. The system uses the Apriori algorithm or the FP-Growth algorithm to mine frequent item sets and association rules from the user access dataset. For example, the system can discover the association rule that "users who visit news pages on weekday mornings usually visit social media pages during lunch breaks", thereby identifying the user's access behavior patterns in different business scenarios.

Through cluster analysis and association rule mining, the system can determine the changing trend of user access behavior. The changing trend of access behavior reflects the access rules of users in different time periods and different behavior patterns, which helps predict future user behavior. For example, the system can predict the changing trend of the access volume of a group of users in a specific period of time based on the user's access behavior characteristics and association rules, thereby providing a basis for resource scheduling and business optimization.

This embodiment solves the limitation problem of traditional user behavior analysis methods in dealing with massive data and dynamic changes. Through cluster analysis and association rule mining, the system can deeply explore the inherent laws and associations in user behavior data, and provide accurate support for behavior prediction and personalized services. For example, after identifying the user's access behavior characteristics, the system can provide customized business recommendations and optimization services for high-frequency access users; after discovering the association between user behaviors, the system can optimize resource allocation for different business scenarios to improve the overall service quality.

The technical effect of this embodiment is significant. Through cluster analysis and association rule mining, the system can maintain efficient and accurate analysis capabilities in user behavior analysis and prediction. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively understand and predict user behavior, and maintain high accuracy even in the face of complex and changeable user access patterns. For example, when the system analyzes real-time traffic data, no matter how complex the user behavior is, the system can accurately identify and predict the changing trend of user behavior through cluster analysis and association rule mining, so as to adjust business strategies in a timely manner and improve user experience and satisfaction.

In general, step S102 solves the complexity and dynamic change problems in user behavior analysis through cluster analysis and association rule mining. Through the collaborative work of cluster analysis, feature extraction and association rule mining, the system not only improves the accuracy of user behavior analysis, but also ensures the efficiency of behavior prediction and personalized services. In the practice of user behavior analysis on the cloud platform, this technical solution significantly improves the efficiency of analysis and decision-making, and effectively improves user satisfaction and business operation efficiency.

Step S103: Input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and determine whether there is a traffic peak or abnormal event based on the output of the decision tree regression prediction model. When the traffic peak or abnormal event is detected, dynamically adjust the resource allocation of the cloud platform system based on the traffic change demand output by the decision tree regression prediction model, and release excess resources after the real-time traffic data of the cloud platform system drops back to normal levels until the standard configuration is restored.

Optionally, in this embodiment, step S103 involves inputting the traffic change trend and the access behavior change trend into a set decision tree regression prediction model. The output of the model is used to determine whether there is a traffic peak or abnormal event, and dynamically adjust resource allocation based on the model output. After the traffic drops back to normal levels, the system releases excess resources until the standard configuration is restored. The technical process of this step includes data input, model prediction, resource adjustment and recovery, aiming to improve the resource management efficiency and responsiveness of the cloud platform.

First, in step S103, the system inputs the traffic change trend and access behavior change trend determined in steps S101 and S102 into the set decision tree regression prediction model. Decision tree regression is a machine learning algorithm that iteratively segments data through a tree structure to find the best prediction result. The system pre-trains a decision tree regression model, which is based on historical traffic data and user behavior data to learn the relationship between different feature combinations and traffic peaks or abnormal events. During the training process, the system uses labeled historical data sets, continuously segments data through the decision tree algorithm, and adjusts model parameters to improve prediction accuracy.

After inputting the traffic change trend and access behavior change trend into the model, the system determines whether there is a traffic peak or abnormal event based on the output of the decision tree regression prediction model. The model output includes the traffic forecast value for a period of time in the future and the probability value of abnormal events. For example, the model may predict that the traffic will increase significantly in the next hour and give the probability of a traffic peak. Based on these prediction results, the system determines whether preventive measures need to be taken at the moment.

When a traffic peak or abnormal event is detected, the system dynamically adjusts the resource allocation of the cloud platform system according to the traffic change demand output by the decision tree regression prediction model. Specifically, the system can automatically expand the number of server instances, increase bandwidth, adjust load balancing strategies, etc. For example, if the model predicts that traffic will increase to twice the current level within the next hour, the system can start additional computing resources in advance to cope with the upcoming traffic peak. This ensures that during traffic peaks, user requests can be processed in a timely manner to avoid performance degradation or service interruptions due to insufficient resources.

After the traffic peak or abnormal event ends, the system continuously monitors real-time traffic data, and after the traffic drops back to normal levels, it releases excess resources until the standard configuration is restored. This step is intended to optimize resource usage and reduce operating costs. For example, when the system detects that the traffic has dropped back to normal levels and predicts that there will be no more abnormal peaks in the future, the system can gradually shut down computing instances that are no longer needed, adjust bandwidth configuration, and return to normal operation. This embodiment not only ensures the stable operation of the system during peak periods, but also effectively avoids waste of resources.

This embodiment solves the lag and inefficiency of traditional resource management methods when dealing with sudden traffic and dynamic changes. Through real-time prediction and dynamic adjustment, the system can identify and respond to traffic peaks and abnormal events in advance, ensuring the timeliness and effectiveness of resource allocation. For example, during a large-scale online event, the cloud platform can use this technical solution to predict and dynamically adjust resource allocation in advance, ensuring stable service operation during the event and avoiding system crashes caused by traffic surges.

The technical effect of this embodiment is significant. Through the decision tree regression prediction model, the system can maintain efficient and accurate prediction and adjustment capabilities in traffic prediction and resource management. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and predict traffic changes, and maintain high accuracy even in the face of complex and changeable traffic patterns. For example, when the system analyzes real-time traffic data, no matter how complex the traffic changes, the system can accurately predict future traffic trends through the decision tree regression model, thereby adjusting resource allocation in a timely manner and improving the stability and performance of the overall system.

In general, step S103 solves the problem of dynamic adjustment and resource optimization in traffic management by applying the decision tree regression prediction model. Through the collaborative work of data input, model prediction, resource adjustment and recovery, the system not only improves the accuracy of traffic prediction, but also ensures the efficiency and flexibility of resource management. In the resource management practice of the cloud platform, this technical solution significantly improves the efficiency of prediction and decision-making, effectively reduces operational risks, and ensures the stable operation of the system.

From the above description, it can be seen that the cloud platform system traffic dynamic balancing processing method provided in the embodiment of the present application can monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend; perform cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data to determine the user's access behavior characteristics, and determine the corresponding access behavior change trend based on the access behavior characteristics and through a preset association rule mining algorithm; input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and when a traffic peak or abnormal event is detected, dynamically adjust the resource allocation of the cloud platform system, thereby effectively improving the flexibility and response speed of the system, optimizing resource utilization, and ensuring the stable operation of the system under peak traffic.

In one embodiment of the cloud platform system traffic dynamic balancing processing method of the present application, referring to FIG. 2 , the following contents may also be specifically included:

Step S201: collecting user access traffic data from the cloud platform system in real time, cleaning and converting the user access traffic data, and converting the user access traffic data into a time series format;

Step S202: determining the optimal parameter combination with the smallest prediction error of the ARIMA time series model on the user access flow data converted into a time series format through grid search, and determining the flow change trend according to the output of the ARIMA time series model.

Optionally, in this embodiment, step S201 and step S202 respectively involve real-time collection and processing of user access traffic data, and traffic forecasting using the ARIMA time series model. The specific technical process includes data collection, data cleaning and conversion, model parameter optimization, and traffic trend forecasting. The collaborative work of these two steps aims to improve the accuracy and efficiency of traffic forecasting and solve the shortcomings of traditional methods in large-scale data processing and dynamic forecasting.

First, in step S201, the system collects user access traffic data from the cloud platform in real time. This data includes the time of user access, IP address, requested resource type, access frequency, etc. The data collection module is connected to the traffic monitoring system of the cloud platform, and uses traffic collection tools (such as Fluentd, Logstash) to collect user access logs in real time and transmit them to the data processing module for further processing.

Next, the system cleans and converts the collected user access traffic data. The purpose of data cleaning is to remove noise and irrelevant information to ensure data quality. Common cleaning steps include deduplication, processing missing values and outliers, and standardizing field formats. For example, the system can remove duplicate access records, fill in missing timestamps, or identify and process abnormal high-frequency access records through statistical methods. After data cleaning is completed, the system converts the cleaned traffic data into a time series format. The time series format means that the data is arranged in chronological order, and each time point corresponds to one or more observations. Through this conversion, the system can retain the time dependency of the data, which is convenient for subsequent time series analysis.

In step S202, the system uses a grid search method to determine the optimal parameter combination of the ARIMA time series model on the user access traffic data converted into a time series format. The ARIMA model (autoregressive integrated moving average model) is a commonly used time series prediction method that models time series data through three parts: autoregression, difference, and moving average. In order to find the optimal parameter combination (p, d, q) of the ARIMA model, the system uses a grid search method to search in a predefined parameter space. The grid search performs an exhaustive search for different parameter combinations, and uses a cross-validation method to calculate the prediction error of each combination, and finally selects the parameter combination with the smallest prediction error. For example, the system searches for different p (autoregressive order), d (difference order), and q (moving average order) in the parameter space, and calculates the mean square error (MSE) of each combination on the validation data set, and selects the combination with the smallest MSE as the optimal parameter.

Once the optimal parameter combination is determined, the system determines the traffic change trend based on the output of the ARIMA time series model. After the model training is completed, the system uses the model to predict future user access traffic and generate a traffic change trend chart. For example, the system can predict daily access traffic in the next week and adjust resource allocation strategies based on the prediction results to cope with possible traffic fluctuations.

This embodiment solves the limitations of traditional traffic prediction methods in large-scale data processing and dynamic change response. Through real-time data collection and cleaning, the system can ensure the accuracy and timeliness of the data; through time series conversion and ARIMA model optimization, the system can capture the time dependency in the data and improve the accuracy and reliability of the prediction. For example, during an e-commerce promotion event, the cloud platform was able to predict in advance and dynamically adjust resource allocation through this technical solution, ensuring the stable operation of services during the event and avoiding system crashes caused by traffic surges.

The technical effect of this embodiment is significant. Through real-time data processing and optimized ARIMA models, the system can maintain efficient and accurate prediction capabilities in traffic forecasting. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and predict traffic changes, and maintain high accuracy even in the face of complex and changeable traffic patterns. For example, when the system analyzes real-time traffic data, no matter how complex the traffic changes, the system can accurately predict future traffic trends through the ARIMA model, thereby adjusting resource allocation in a timely manner and improving the stability and performance of the overall system.

In general, step S201 and step S202 solve the data processing and model optimization problems in traffic prediction through the collaborative work of data collection, cleaning and conversion, model parameter optimization and trend prediction. Through efficient data processing and accurate model prediction, the system not only improves the accuracy of traffic prediction, but also ensures efficient and flexible resource management. In the practice of traffic prediction on the cloud platform, this technical solution significantly improves the efficiency of prediction and decision-making, effectively reduces operational risks, and ensures the stable operation of the system.

In one embodiment of the cloud platform system traffic dynamic balancing processing method of the present application, referring to FIG. 3 , the following contents may also be specifically included:

Step S301: Acquire a user access pattern and a user access service corresponding to the real-time traffic data;

Step S302: Divide the users into different types of groups according to the user access patterns, the user access services and a preset K-Means clustering algorithm, and determine the user access behavior characteristics, wherein each group has unique access behavior characteristics.

Optionally, in this embodiment, step S301 and step S302 respectively involve obtaining user access patterns and services, and classifying users using the K-Means clustering algorithm. The collaboration of these two steps is intended to identify access characteristics of different user groups, thereby improving the system's responsiveness and optimizing resource allocation.

First, in step S301, the system extracts information related to user access from real-time traffic data. User access patterns include user access frequency, access time period, access duration, access path, etc., while user access services involve specific operations performed by users on the platform, such as browsing products, adding shopping carts, placing orders and paying, etc. The system collects this data in real time and performs preliminary processing by combining log analysis tools and user behavior analysis platforms (such as Google Analytics and Mixpanel). For example, the system can extract each user's IP address, access timestamp, request URL and other information from the access log, and associate this information with the user's operation records on the platform to form a complete user access record.

Next, in step S302, the system uses the preset K-Means clustering algorithm to divide users into different types of groups according to their access patterns and access services. K-Means clustering is an unsupervised learning algorithm that divides data points into K clusters by minimizing the square error within the group. In the specific implementation process, the system first extracts features and normalizes the user access data, such as converting the user's access frequency, access duration, number of business operations, etc. into numerical features and normalizing them. Then, the system inputs these feature vectors into the K-Means algorithm, sets the initial number of clusters K, and the algorithm finds the optimal cluster center and user group division through iterative optimization. For example, for an e-commerce platform, the system can set K=5 and divide users into five groups, such as high-frequency purchasing users, browsing users, occasional purchasing users, etc.

Through K-Means clustering, the system can determine the access behavior characteristics of each group. Each group has unique access patterns and business operation habits. For example, high-frequency purchasing users may show high access frequency, short access intervals, and frequent ordering operations, while browsing users may show long page stays and fewer purchase operations. By analyzing these characteristics, the system formulates differentiated service strategies and resource allocation plans for different user groups. For example, for high-frequency purchasing users, the system can prioritize the allocation of more computing resources and bandwidth to ensure their access experience; for browsing users, the system can improve their conversion rate through personalized recommendations and marketing strategies.

This embodiment solves the shortcomings of traditional user behavior analysis methods in processing large-scale user data and dynamic behavior identification. Through real-time data collection and K-Means clustering, the system can quickly identify the behavioral characteristics of different user groups and improve the efficiency and accuracy of data analysis. For example, during a large-scale promotional event, the system can use this technical solution to identify high-frequency purchasing users and potential purchasing users in advance, and push preferential information and resources in a targeted manner, significantly improving the conversion rate and user satisfaction of the event.

The technical effect of this embodiment is significant. Through real-time user behavior analysis and optimized clustering algorithms, the system can maintain efficient and accurate analysis capabilities in user classification and behavior identification. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and analyze user behavior, and maintain high accuracy even in the face of complex and changeable user access patterns. For example, when the system analyzes real-time user data, no matter how complex the user behavior is, the system can accurately identify the characteristics of different user groups through K-Means clustering, thereby formulating more accurate marketing and resource allocation strategies, and improving the overall system's responsiveness and user satisfaction.

In general, step S301 and step S302 solve the data processing and dynamic identification problems in user behavior analysis through the collaborative work of data collection, behavior analysis, and clustering algorithms. Through efficient data processing and precise clustering analysis, the system not only improves the accuracy of user classification and behavior identification, but also ensures efficient and flexible resource management. In the practice of user behavior analysis on the cloud platform, this technical solution significantly improves the efficiency of analysis and decision-making, effectively reduces operational risks, and ensures the stable operation of the system.

In one embodiment of the cloud platform system traffic dynamic balancing processing method of the present application, referring to FIG. 4 , the following contents may also be specifically included:

Step S401: determining the access behaviors of different user groups based on user access behavior feature analysis;

Step S402: using the Apriori association rule mining algorithm to determine the correlation between the access behaviors of different user groups in their respective business scenarios on the user access data set, and determining the corresponding access behavior change trend according to the correlation.

Optionally, in this embodiment, step S401 and step S402 respectively involve analyzing the access behaviors of different user groups according to the user access behavior characteristics, and determining the correlation between these behaviors using the Apriori association rule mining algorithm. Through these two steps, the system can deeply understand the user behavior patterns, identify the access behavior change trends of different user groups in their respective business scenarios, and thus optimize resource allocation and user experience.

First, in step S401, the system analyzes the detailed access behavior characteristics of different user groups previously divided by the K-Means clustering algorithm. These characteristics include the user's access frequency, access time period, access path, residence time, and specific business operations (such as product browsing, shopping cart operations, order submission, etc.). By analyzing these behavioral characteristics, the system can identify the typical behavior patterns of each user group. For example, high-frequency purchasing users may mainly visit frequently and make multiple purchases within a specific time period; while browsing users may stay on the product details page for a long time in multiple time periods, but have fewer actual purchase behaviors. By analyzing these characteristics, the system can build a behavioral feature model for each user group and provide basic data for subsequent association rule mining.

Next, in step S402, the system uses the Apriori association rule mining algorithm to determine the correlation between the access behaviors of different user groups in their respective business scenarios on the user access data set. The Apriori algorithm is a classic association rule mining algorithm that aims to discover frequently occurring item sets and their association rules in a data set. In the specific implementation process, the system first converts the user's access behavior into a transaction data set, and each transaction represents a sequence of access behaviors of a user in a specific time period. Then, the system applies the Apriori algorithm to mine frequent item sets and association rules in these transaction data sets. For example, the system can find that high-frequency purchasing users usually add products to the shopping cart and complete the purchase in a short time after visiting the product details page; or browsing users may stay on a specific type of product for a long time after visiting multiple product details pages but ultimately do not place an order.

Through association rule mining, the system can identify the typical behavior patterns and changing trends of different user groups in their respective business scenarios. For example, the system can discover the changing trends of the purchasing behavior of high-frequency purchasing users during price fluctuations or promotional activities, or the changes in the attention and access paths of browsing users on specific types of goods (such as electronic products and clothing). Based on these association rules and behavioral trend analysis, the system can formulate personalized marketing strategies and resource allocation plans for different user groups. For example, for high-frequency purchasing users, the system can accurately push during promotional activities and provide limited-time discounts; for browsing users, the system can improve their purchase conversion rate through personalized recommendations.

This embodiment solves the shortcomings of traditional user behavior analysis methods in dealing with complex behavior patterns and dynamic changes. Through detailed behavior feature analysis and association rule mining, the system can deeply understand the behavior patterns and change trends of different user groups, and improve the depth and accuracy of data analysis. For example, in a large-scale promotional event, the system can use this technical solution to identify in advance which user groups are most interested in which products, and push preferential information and resource allocation in a targeted manner, significantly improving the conversion rate and user satisfaction of the event.

The technical effect of this embodiment is significant. Through detailed behavioral feature analysis and optimized association rule mining algorithms, the system can maintain efficient and accurate analysis capabilities in user behavior pattern recognition and change trend analysis. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and analyze user behavior, and maintain high accuracy even in the face of complex and changeable user access patterns. For example, when the system analyzes real-time user data, no matter how complex the user behavior is, the system can accurately identify the behavioral associations and change trends of different user groups through the Apriori algorithm, thereby formulating more accurate marketing and resource allocation strategies, and improving the overall system's responsiveness and user satisfaction.

In general, step S401 and step S402 solve the problem of complex behavior pattern recognition and dynamic change prediction in user behavior analysis through the collaborative work of behavior feature analysis and association rule mining. Through efficient data processing and precise behavior pattern analysis, the system not only improves the accuracy of user behavior recognition, but also ensures efficient and flexible resource management. In the practice of user behavior analysis on the cloud platform, this technical solution significantly improves the efficiency of analysis and decision-making, effectively reduces operational risks, and ensures the stable operation of the system.

In one embodiment of the cloud platform system traffic dynamic balancing processing method of the present application, referring to FIG5 , the following contents may also be specifically included:

Step S501: training a decision tree regression prediction model according to historical sample data until an optimal prediction rule is formed, and inputting the traffic change trend and the access behavior change trend into a set decision tree regression prediction model;

Step S502: Determine whether there is a traffic peak or abnormal event based on the comparison relationship between the predicted probability value output by the decision tree regression prediction model and the value of a preset threshold.

Optionally, in this embodiment, step S501 and step S502 respectively involve training a decision tree regression prediction model and using the model to determine whether there is a traffic peak or an abnormal event. Through these two steps, the system can accurately predict future traffic changes and promptly identify potential traffic peaks or abnormal events, thereby optimizing system resource allocation and improving user experience.

First, in step S501, the system trains a decision tree regression prediction model based on historical sample data. Decision tree regression is a commonly used machine learning algorithm that can predict continuous values by building a tree model. In the specific implementation process, the system first collects and organizes historical traffic data and user access behavior data, which include user visits, page clicks, access paths, dwell time, and business operation records in different time periods. Then, the system preprocesses these data, such as data cleaning, feature extraction, and standardization, to ensure data quality and model training effect.

After data preprocessing is completed, the system inputs the processed data set into the decision tree regression algorithm for training. During the training process, the system continuously splits the data set and optimizes the splitting conditions of each node to minimize the prediction error and eventually form the optimal prediction rule. For example, the system can use the decision tree regression model to predict the number of visits in the next hour, and determine the trend of the number of visits based on the access patterns and business operations in the historical data. When the model training is completed, the system inputs the traffic change trend and the access behavior change trend into the trained decision tree regression model for prediction, and generates the predicted value of traffic change in the future period of time.

Next, in step S502, the system determines whether there is a traffic peak or abnormal event based on the comparison relationship between the predicted probability value output by the decision tree regression prediction model and the value of the preset threshold. Specifically, the system compares the predicted traffic change value with the set traffic threshold. If the predicted value exceeds the threshold, it is determined that there may be a traffic peak or abnormal event. For example, the system can set a traffic threshold. If the predicted access volume exceeds the threshold, the system will trigger an alarm in the background and take corresponding countermeasures, such as increasing server resources, adjusting the load balancing strategy, etc.

This embodiment solves the shortcomings of traditional traffic prediction methods in processing large-scale data and dynamic changes. By training the decision tree regression model with historical data, the system can accurately predict future traffic changes based on user access behavior and traffic change trends, and promptly identify potential traffic peaks or abnormal events. For example, on the eve of a large-scale promotional event, the system can use this technical solution to predict the traffic peak during the event in advance, and increase server resources and bandwidth configuration in advance to ensure stable operation of the system and a good user experience during the event.

The technical effect of this embodiment is significant. Through an efficient decision tree regression prediction model and an accurate traffic peak judgment algorithm, the system can maintain efficient and accurate analysis capabilities in traffic prediction and abnormal event identification. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and predict user traffic, and maintain high accuracy even in the face of complex and changeable user access patterns. For example, when the system analyzes real-time user data, no matter how complex the traffic changes, the system can accurately predict future traffic changes through a decision tree regression model, and promptly identify potential traffic peaks and abnormal events, thereby formulating more accurate resource allocation and response strategies, and improving the overall system's responsiveness and user satisfaction.

In general, step S501 and step S502 solve the data processing and dynamic change problems in traffic prediction and abnormal event identification through the collaborative work of historical data training, decision tree regression prediction and traffic peak judgment. Through efficient data processing and accurate prediction analysis, the system not only improves the accuracy of traffic prediction, but also ensures efficient and flexible resource management. In the practice of traffic prediction on the cloud platform, this technical solution significantly improves the efficiency of analysis and decision-making, effectively reduces operational risks, and ensures the stable operation of the system.

In one embodiment of the cloud platform system traffic dynamic balancing processing method of the present application, referring to FIG. 6 , the following contents may also be specifically included:

Step S601: when a traffic peak or an abnormal event is detected, the key factors causing the abnormality are determined according to the output of the decision tree regression prediction model;

Step S602: Dynamically adjust the resource allocation strategy of the cloud platform system according to the traffic change demand predicted by the decision tree regression model and the key factors causing the anomaly.

Optionally, in this embodiment, step S601 and step S602 respectively involve determining the key factors causing the anomaly when detecting a traffic peak or an abnormal event, and dynamically adjusting the resource allocation strategy according to the predicted traffic change demand and the key factors. Through these two steps, the system can identify and respond to traffic anomalies in a timely manner, ensuring stable operation of the system and improving user experience.

First, in step S601, when the system detects the presence of a traffic peak or an abnormal event, the key factors causing the abnormality are determined based on the output of the decision tree regression prediction model. In the specific implementation process, the system will use the previously trained decision tree regression model to analyze the current and historical traffic data. When the predicted value output by the model exceeds the preset threshold, indicating that a traffic peak or an abnormal event may have occurred, the system will start the abnormal cause analysis module. This module reversely parses the decision tree model to find out the main features and nodes that affect the prediction results. For example, if the system detects a sudden surge in visits within a certain period of time, the model may indicate that this is caused by concentrated visits by a specific user group, the launch of a promotional activity, or abnormal use of a business function.

By identifying the key factors of the anomaly, the system can more accurately understand the root cause of the traffic anomaly. For example, during a promotion, the system may find that the surge in traffic is due to a large number of users visiting a specific product page in a short period of time, or because users frequently refresh the page when they encounter problems on the checkout page. After identifying these key factors, the system can take more targeted measures to solve potential problems and reduce unnecessary waste of resources.

Next, in step S602, the system dynamically adjusts the resource allocation strategy of the cloud platform system according to the traffic change demand predicted by the decision tree regression model and the key factors causing the anomaly. Specifically, the system will adjust server resources, network bandwidth, cache strategy, etc. in real time according to the predicted traffic demand and identified key factors. For example, if it is predicted that there will be a peak in access within a certain time period, the system can increase the corresponding server instances in advance and expand the network bandwidth to ensure smooth user access and quick response. If a business function is detected to have an anomaly that causes a surge in traffic, the system can temporarily adjust the service strategy of the function, limit its access frequency or optimize its processing flow to reduce the system load.

This embodiment solves the lag and inflexibility of traditional resource allocation methods when facing sudden traffic changes. Through real-time monitoring and dynamic adjustment, the system can quickly respond and adjust resource allocation when traffic peaks or abnormal events occur, ensuring the stability and efficient operation of the system. For example, during a sudden user access peak, the system can increase server resources and network bandwidth in a timely manner through this technical solution, prevent system overload and service interruption, and ensure a good user experience.

The technical effect of this embodiment is significant. Through precise anomaly analysis and dynamic resource adjustment, the system can maintain efficient and flexible response capabilities in terms of traffic management and resource optimization. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and respond to traffic changes, and maintain high accuracy even in the face of complex and changeable user access patterns. For example, when the system analyzes real-time user data, no matter how complex the traffic changes, the system can accurately predict future traffic demands through decision tree regression models and dynamic adjustment strategies, and adjust resource allocation in a timely manner to improve the overall system's responsiveness and user satisfaction.

In general, step S601 and step S602 solve the data processing and resource allocation problems in traffic management through the collaborative work of anomaly detection, key factor analysis and dynamic resource adjustment. Through efficient data processing and accurate predictive analysis, the system not only improves the accuracy of traffic management, but also ensures the flexibility and efficiency of resource management. In the practice of traffic management on the cloud platform, this technical solution significantly improves the efficiency of analysis and decision-making, effectively reduces operational risks, and ensures the stable operation of the system.

In one embodiment of the cloud platform system traffic dynamic balancing processing method of the present application, referring to FIG. 7 , the following contents may also be specifically included:

Step S701: monitoring the real-time traffic data of the cloud platform system;

Step S702: After monitoring the real-time traffic data of the cloud platform system and returning to a normal level, gradually reduce the number of computing nodes and storage space until the resource configuration of the system is completely restored to a pre-set standard state.

Optionally, in this embodiment, step S701 and step S702 respectively involve monitoring the real-time traffic data of the system, and gradually reducing the number of computing nodes and storage space after the traffic data falls back to a normal level until the resource configuration returns to a standard state. Through these two steps, the system can effectively manage resources, ensure stable operation during peak traffic periods, and optimize resource utilization after the traffic falls back.

First, in step S701, the system monitors real-time traffic data. In the specific implementation process, the system will deploy multiple monitoring nodes, which are distributed on different servers and network devices to collect key indicators such as user visits, page clicks, request response time, and bandwidth usage in real time. The system uses the data acquisition module to aggregate these real-time data to the central monitoring platform for unified analysis and processing. The technical principle of real-time traffic monitoring mainly relies on data acquisition, transmission and analysis technology to ensure the real-time and accuracy of data. For example, the system can collect traffic data once a second, and use the traffic monitoring algorithm to analyze data fluctuations in real time and identify potential traffic peaks or abnormal events.

Through real-time traffic monitoring, the system can promptly detect and respond to traffic changes, ensuring that sufficient computing and storage resources are provided during peak traffic periods. For example, during a promotional event, the system discovered through real-time monitoring that user traffic increased rapidly. The system can immediately start the automatic expansion mechanism to increase computing nodes and storage space to prevent system overload and service interruption. At the same time, real-time monitoring can also help the system identify potential security threats, such as DDoS attacks, and ensure the security and stability of the system through timely warnings and the initiation of protective measures.

Next, in step S702, when the real-time traffic data is monitored to fall back to normal levels, the system gradually reduces the number of computing nodes and storage space until the resource configuration is fully restored to the pre-set standard state. Specifically, the system will gradually shut down or release some computing nodes and storage resources according to the changing trend of real-time traffic data. For example, if the system monitors that the user access volume is gradually decreasing, the system will first shut down or release the computing nodes with lower loads according to the preset resource reduction strategy, gradually reduce bandwidth and storage allocation, and ensure efficient use of resources.

This embodiment solves the lag and resource waste problems of traditional resource management methods when facing dynamic traffic changes. Through real-time monitoring and dynamic adjustment, the system can provide sufficient resource support during traffic peaks, and reduce resource allocation in time after traffic drops to avoid resource waste. For example, after a large-scale promotional event, the system can accurately identify the time point when traffic drops through this technical solution, and gradually reduce computing nodes and storage space to ensure the optimal configuration and efficient use of system resources.

The technical effect of this embodiment is significant. Through accurate real-time traffic monitoring and dynamic resource adjustment, the system can maintain efficient and flexible response capabilities in resource management and optimization. Specifically, after implementing this technical solution, a certain cloud platform can more comprehensively monitor and respond to traffic changes, and maintain efficient resource utilization regardless of traffic peaks or declines. For example, when the system analyzes real-time user data, no matter how complex the traffic changes, the system can accurately identify traffic change trends through real-time monitoring and dynamic adjustment strategies, and adjust resource allocation in a timely manner, thereby improving the overall system's responsiveness and user satisfaction.

In general, step S701 and step S702 solve the problems of dynamic changes and resource optimization in traffic management through the collaborative work of real-time traffic monitoring and dynamic resource adjustment. Through efficient data processing and precise monitoring and analysis, the system not only improves the accuracy of traffic management, but also ensures the flexibility and efficiency of resource management. In the traffic management practice of the cloud platform, this technical solution has significantly improved the efficiency of analysis and decision-making, effectively reduced operational risks, and ensured the stable operation of the system. After the real-time traffic data of the platform system falls back to normal levels, the number of computing nodes and storage space are gradually reduced until the system's resource configuration is fully restored to the pre-set standard state.

Optionally, in this embodiment,

In order to effectively improve the flexibility and response speed of the system, optimize resource utilization, and ensure the stable operation of the system under peak traffic, the present application provides an embodiment of a cloud platform system traffic dynamic balancing processing device for implementing all or part of the content of the cloud platform system traffic dynamic balancing processing method. Referring to FIG. 8 , the cloud platform system traffic dynamic balancing processing device specifically includes the following content:

The flow change determination module 10 is used to monitor the real-time flow data in the cloud platform system, perform time series model analysis on the real-time flow data within a set time period, and determine the corresponding flow change trend;

The access behavior determination module 20 is used to perform cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data, determine the user's access behavior characteristics according to the results of the cluster analysis, and determine the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the user's access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm;

The abnormal prediction module 30 is used to input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and determine whether there is a traffic peak or an abnormal event based on the output of the decision tree regression prediction model. When the traffic peak or abnormal event is detected, the resource allocation of the cloud platform system is dynamically adjusted according to the traffic change demand output by the decision tree regression prediction model, and after the real-time traffic data of the cloud platform system drops back to a normal level, excess resources are released until the standard configuration is restored.

From the above description, it can be seen that the cloud platform system traffic dynamic balancing processing device provided in the embodiment of the present application can monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend; perform cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data to determine the user's access behavior characteristics, and determine the corresponding access behavior change trend based on the access behavior characteristics and through a preset association rule mining algorithm; input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and when a traffic peak or abnormal event is detected, dynamically adjust the resource allocation of the cloud platform system, thereby effectively improving the flexibility and response speed of the system, optimizing resource utilization, and ensuring the stable operation of the system under peak traffic.

From the hardware level, in order to effectively improve the flexibility and response speed of the system, optimize resource utilization, and ensure the stable operation of the system under peak traffic, the present application provides an embodiment of an electronic device for implementing all or part of the content of the cloud platform system traffic dynamic balancing processing method, and the electronic device specifically includes the following content:

Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the cloud platform system traffic dynamic balancing processing device and the core business system, user terminal and related database and other related devices; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the cloud platform system traffic dynamic balancing processing method and the embodiment of the cloud platform system traffic dynamic balancing processing device in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

In practical applications, part of the cloud platform system traffic dynamic balancing processing method can be executed on the electronic device side as described above, or all operations can be completed in the client device. The specific selection can be based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

FIG9 is a schematic block diagram of a system structure of an electronic device 9600 according to an embodiment of the present application. As shown in FIG9 , the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that FIG9 is exemplary; other types of structures may also be used to supplement or replace the structure to implement telecommunication functions or other functions.

In one embodiment, the cloud platform system traffic dynamic balancing processing method function may be integrated into the central processor 9100. The central processor 9100 may be configured to perform the following control:

Step S101: monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend;

Step S102: performing cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data, determining the user's access behavior characteristics according to the results of the cluster analysis, and determining the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the user's access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm;

Step S103: Input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and determine whether there is a traffic peak or abnormal event based on the output of the decision tree regression prediction model. When the traffic peak or abnormal event is detected, dynamically adjust the resource allocation of the cloud platform system based on the traffic change demand output by the decision tree regression prediction model, and release excess resources after the real-time traffic data of the cloud platform system drops back to normal levels until the standard configuration is restored.

From the above description, it can be seen that the electronic device provided in the embodiment of the present application monitors the real-time traffic data in the cloud platform system, performs time series model analysis on the real-time traffic data within a set time period, and determines the corresponding traffic change trend; performs cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data to determine the user's access behavior characteristics, and determines the corresponding access behavior change trend based on the access behavior characteristics and through a preset association rule mining algorithm; inputs the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and dynamically adjusts the resource allocation of the cloud platform system when a traffic peak or abnormal event is detected, thereby effectively improving the flexibility and response speed of the system, optimizing resource utilization, and ensuring stable operation of the system under peak traffic.

In another embodiment, the cloud platform system traffic dynamic balancing processing device can be configured separately from the central processor 9100. For example, the cloud platform system traffic dynamic balancing processing device can be configured as a chip connected to the central processor 9100, and the cloud platform system traffic dynamic balancing processing method function can be realized through the control of the central processor.

As shown in FIG9 , the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily include all the components shown in FIG9 ; in addition, the electronic device 9600 may also include components not shown in FIG9 , and reference may be made to the prior art.

As shown in FIG. 9 , the central processor 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and/or logic device. The central processor 9100 receives input and controls the operation of various components of the electronic device 9600 .

The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing.

The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application/function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and/or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and/or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

The communication module 9110 is a transmitter/receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter/receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and/or a wireless LAN module, etc. The communication module 9110 (transmitter/receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the method for dynamically balancing the flow of a cloud platform system in the above-mentioned embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the method for dynamically balancing the flow of a cloud platform system in the above-mentioned embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

Step S101: monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend;

Step S102: performing cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data, determining the user's access behavior characteristics according to the results of the cluster analysis, and determining the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the user's access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm;

Step S103: Input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and determine whether there is a traffic peak or abnormal event based on the output of the decision tree regression prediction model. When the traffic peak or abnormal event is detected, dynamically adjust the resource allocation of the cloud platform system based on the traffic change demand output by the decision tree regression prediction model, and release excess resources after the real-time traffic data of the cloud platform system drops back to normal levels until the standard configuration is restored.

From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application monitors the real-time traffic data in the cloud platform system, performs time series model analysis on the real-time traffic data within a set time period, and determines the corresponding traffic change trend; performs cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data to determine the user's access behavior characteristics, and determines the corresponding access behavior change trend based on the access behavior characteristics and through a preset association rule mining algorithm; inputs the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and dynamically adjusts the resource allocation of the cloud platform system when a traffic peak or abnormal event is detected, thereby effectively improving the flexibility and response speed of the system, optimizing resource utilization, and ensuring the stable operation of the system under peak traffic.

The embodiments of the present application also provide a computer program product capable of implementing all the steps of the method for dynamically balancing the flow of a cloud platform system in the above embodiments, where the execution subject is a server or a client. When the computer program/instruction is executed by a processor, the steps of the method for dynamically balancing the flow of a cloud platform system are implemented. For example, the computer program/instruction implements the following steps:

Step S101: monitor the real-time traffic data in the cloud platform system, perform time series model analysis on the real-time traffic data within a set time period, and determine the corresponding traffic change trend;

Step S102: performing cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data, determining the user's access behavior characteristics according to the results of the cluster analysis, and determining the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the user's access behaviors in different business scenarios determined on the user access data set by a preset association rule mining algorithm;

Step S103: Input the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and determine whether there is a traffic peak or abnormal event based on the output of the decision tree regression prediction model. When the traffic peak or abnormal event is detected, dynamically adjust the resource allocation of the cloud platform system based on the traffic change demand output by the decision tree regression prediction model, and release excess resources after the real-time traffic data of the cloud platform system drops back to normal levels until the standard configuration is restored.

From the above description, it can be seen that the computer program product provided in the embodiment of the present application monitors the real-time traffic data in the cloud platform system, performs time series model analysis on the real-time traffic data within a set time period, and determines the corresponding traffic change trend; performs cluster analysis on the user access patterns and user access services corresponding to the real-time traffic data to determine the user's access behavior characteristics, and determines the corresponding access behavior change trend based on the access behavior characteristics and through a preset association rule mining algorithm; inputs the traffic change trend and the access behavior change trend into a set decision tree regression prediction model, and dynamically adjusts the resource allocation of the cloud platform system when a traffic peak or abnormal event is detected, thereby effectively improving the flexibility and response speed of the system, optimizing resource utilization, and ensuring the stable operation of the system under peak traffic.

It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

The present invention is described with reference to the flowchart and/or block diagram of the method, device (apparatus), and computer program product according to the embodiment of the present invention. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the processes and/or boxes in the flowchart and/or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the functions specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims (10)

1. The method for dynamically balancing and processing the flow of the cloud platform system is characterized by comprising the following steps of:

monitoring real-time flow data in a cloud platform system, performing time sequence model analysis on the real-time flow data in a set time period, and determining a corresponding flow change trend;

Performing cluster analysis on a user access mode and a user access service corresponding to the real-time flow data, determining access behavior characteristics of the user according to the result of the cluster analysis, and determining corresponding access behavior change trend according to the access behavior characteristics and the correlation between the access behaviors of the user determined on a user access data set through a preset association rule mining algorithm under different service scenes;

And inputting the flow change trend and the access behavior change trend into a set decision tree regression prediction model, judging whether a flow peak or an abnormal event exists according to the output of the decision tree regression prediction model, dynamically adjusting the resource allocation of the cloud platform system according to the flow change demand output by the decision tree regression prediction model when the flow peak or the abnormal event is detected, and releasing redundant resources until standard configuration is restored after the real-time flow data of the cloud platform system fall back to a normal level.

2. The method for dynamically balancing flow of a cloud platform system according to claim 1, wherein the monitoring real-time flow data in the cloud platform system, performing time-series model analysis on the real-time flow data in a set period of time, and determining a corresponding flow change trend, includes:

Collecting user access flow data in real time from a cloud platform system, cleaning and converting the user access flow data, and converting the user access flow data into a time sequence format;

And determining an optimal parameter combination with the minimum prediction error of the ARIMA time sequence model on the user access flow data converted into the time sequence format through grid search, and determining the flow change trend according to the output of the ARIMA time sequence model.

3. The method for dynamically balancing the flow of the cloud platform system according to claim 1, wherein the performing cluster analysis on the user access mode and the user access service corresponding to the real-time flow data, and determining the access behavior feature of the user according to the result of the cluster analysis, includes:

acquiring a user access mode and a user access service corresponding to the real-time flow data;

and dividing the users into groups of different types according to the user access mode, the user access service and a preset K-Means clustering algorithm, and determining the access behavior characteristics of the users, wherein each group has unique access behavior characteristics.

4. The method for dynamically balancing the flow of the cloud platform system according to claim 1, wherein the determining the corresponding trend of the access behavior according to the access behavior characteristics and the correlation between the access behaviors of the user determined on the user access data set by the preset association rule mining algorithm under different service scenarios comprises:

Determining access behaviors of different user groups according to the user access behavior feature analysis;

and determining the correlation between the access behaviors of different user groups in respective service scenes on the user access data set by using an Apriori association rule mining algorithm, and determining the corresponding access behavior change trend according to the correlation.

5. The method for dynamically balancing flow of a cloud platform system according to claim 1, wherein the step of inputting the flow variation trend and the access behavior variation trend into a decision tree regression prediction model, and determining whether a flow peak or an abnormal event exists according to an output of the decision tree regression prediction model, comprises:

Training a decision tree regression prediction model according to the historical sample data until an optimal prediction rule is formed, and inputting the flow change trend and the access behavior change trend into a set decision tree regression prediction model;

Judging whether a flow peak or an abnormal event exists according to the comparison relation between the predictive probability value output by the decision tree regression predictive model and the value of a preset threshold value.

6. The method for dynamically balancing flow of the cloud platform system according to claim 1, wherein when the flow peak or the abnormal event is detected, dynamically adjusting the resource allocation of the cloud platform system according to the flow change demand output by the decision tree regression prediction model, comprises:

when detecting that a flow peak or an abnormal event exists, determining key factors causing the abnormality according to the output of the decision tree regression prediction model;

And dynamically adjusting the resource allocation strategy of the cloud platform system according to the flow change requirement predicted by the decision tree regression model and the key factors causing the abnormality.

7. The method for dynamically balancing the flow of the cloud platform system according to claim 1, wherein the step of releasing the redundant resources until the standard configuration is restored after the real-time flow data of the cloud platform system falls back to the normal level comprises the steps of:

monitoring real-time flow data of the cloud platform system;

And gradually reducing the number of the computing nodes and the storage space after monitoring that the real-time flow data of the cloud platform system fall back to the normal level until the resource allocation of the system is completely restored to a preset standard state.

8. A cloud platform system flow dynamic balance processing device, the device comprising:

The flow change determining module is used for monitoring real-time flow data in the cloud platform system, performing time sequence model analysis on the real-time flow data in a set time period and determining a corresponding flow change trend;

The access behavior determining module is used for carrying out cluster analysis on the user access mode and the user access business corresponding to the real-time flow data, determining the access behavior characteristics of the user according to the result of the cluster analysis, and determining the corresponding access behavior change trend according to the access behavior characteristics and the correlation between the access behaviors of the user determined on the user access data set through a preset association rule mining algorithm under different business scenes;

The anomaly prediction module is used for inputting the flow change trend and the access behavior change trend into a set decision tree regression prediction model, judging whether a flow peak or an anomaly event exists according to the output of the decision tree regression prediction model, dynamically adjusting the resource allocation of the cloud platform system according to the flow change demand output by the decision tree regression prediction model when the flow peak or the anomaly event is detected, and releasing redundant resources until standard configuration is restored after the real-time flow data of the cloud platform system fall back to a normal level.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the cloud platform system traffic dynamic balancing processing method of any one of claims 1 to 7 when the program is executed by the processor.

10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the cloud platform system traffic dynamic balancing processing method according to any one of claims 1 to 7.