Intelligent detection method and detection system for server exception of hybrid strategy

Abstract

The invention discloses an intelligent detection method for server exception of a hybrid strategy. The method comprises a historical data acquisition and analysis step and a real-time anomaly alarm step, wherein the historical data acquisition and analysis step comprises the steps of acquiring server historical data in a preset time period, preprocessing the historical data, screening out featuresneeding to be analyzed, and taking data corresponding to each feature as a time sequence, determining a normal value and an abnormal value corresponding to each time sequence in the historical data in combination with a time sequence decomposition algorithm and a Grubbs test algorithm; calculating a normal threshold value corresponding to each feature in the historical data, i.e., a maximum valuerange and a minimum value range of a normal value in each time sequence; and taking all time sequences in the historical data as training samples, establishing a plurality of abnormal data detectionmodels to predict and judge the to-be-detected sample, and outputting the probability that the to-be-detected sample is abnormal, so as to analyze the real-time data in the real-time alarm step.

Description

technical field [0001] The invention relates to a mixed-strategy server abnormal intelligent detection method and detection system, belonging to the technical field of computers. Background technique [0002] In recent years, with the continuous improvement of enterprises' requirements for computing power, storage capacity, and network resources, there are more and more platforms that provide servers for customers. However, various abnormalities or failures may occur during the operation of various servers, which will cause many impacts on the actual use of customers and bring inconvenience to the management of the platform. Since the server is operating continuously for a long time, in order to provide stable service quality and provide customers with reliable guarantee, the platform that provides the server needs to monitor the running server in real time, detect and troubleshoot possible abnormalities or Fault. [0003] Server anomaly detection is an important part of s...

AI Technical Summary

Problems solved by technology

Anomaly detection algorithms based on time series are widely used in network traffic detection, financial data analysis, and natural disaster forecasting; however, this algorithm is not suitable for high-dimensional data, and it is difficult to analyze the relationship between server data

Anomaly detection algorithms based on machine learning take advantage of machine learning algorithms in processing large-scale and complex data. Machine learning algorithms have been successfully applied to many fields such as image recognition, network security, and Internet recommendation systems. However, the performance of the model trained by the machine learning algorithm is greatly affected by the training data. Due to the variability and complexity of the server data, it is difficult to directly use the machine learning algorithm to establish a model for abnormal judgment to ensure the accuracy, reliability and reliability of the model. Generalization ability, and the time cost of many machine learning algorithms to build models or perform real-time analysis is too high, which is difficult to meet the needs of practical applications

[0007] On the other hand, judging whether the server is abnormal requires a comprehensive analysis of server data. Due to the complexity of the network environment, the standards for server anomaly detection are not fixed, and need to be adjusted in real time and intelligently according to the actual situation of the server to meet the actual situation. requirements, while the existing server anomaly detection technology lacks consideration of actual scenarios

[0008] In summary, a single technical solution using an anomaly detection algorithm is difficult to apply to complex network environments

In practical applications, the existing server anomaly detection technology lacks flexibility, often cannot achieve good results, and it is difficult to guarantee the accuracy and efficiency of anomaly detection, which cannot meet actual needs

Images