Service anomaly detection method, device, equipment and storage medium

Abstract

The present disclosure relates to a service anomaly detection method, device, device, and storage medium, including: obtaining index values ​​of each reference index of a target service in multiple time intervals in a target historical time period; The preset alarm threshold of the index determines the influence degree of each reference index; based on the index value and the influence degree of the reference index, the machine learning model is trained to obtain the influence degree prediction model; according to the influence degree prediction model, the reference index corresponding to the current moment is obtained. The index value range; if the index value of the reference index at the current moment does not belong to the index value interval, anomaly detection is performed on the target service. Through the impact prediction model, the index value interval corresponding to the index value at the current moment is dynamically determined, so as to determine whether an abnormality occurs based on the interval, and the abnormal situation of the target service can be detected in time, and the stability of the target service operation is improved.

Description

technical field [0001] The present disclosure relates to the technical field of network security, and in particular to a service anomaly detection method, device, equipment and storage medium. Background technique [0002] With the continuous development of Internet technology, related services based on Internet technology have been penetrated into various fields as a new model, such as cloud services that are widely used. In order to ensure the stable operation of cloud services, it is necessary to detect abnormalities of cloud services in time and deal with them. Some abnormal situations are easier to find. For example, the running program of the cloud service has a null pointer exception and other display errors, which can be reported to the administrator for processing in time. However, some abnormal situations are difficult to find. For example, cloud services have abnormal traffic in a certain period of time compared with this period in history. If they cannot be disc...

AI Technical Summary

Problems solved by technology

[0004] The present disclosure provides a service anomaly detection method, device, equipment, and storage medium to at least solve the problem in the related art that by setting a fixed index value alarm interval to discover service operation anomalies, it is easy to cause service anomaly false alarms or missing alarms.

Images