Micro-service fault diagnosis method based on neural network and storage medium

Abstract

The invention provides a micro-service fault diagnosis method based on a neural network. The method comprises the following steps: acquiring a service link diagram; based on the service link diagram, collecting and storing service operation data; analyzing the operation data to obtain indexes influencing service operation; analyzing abnormal fluctuation information in the indexes, and constructing an abnormal index graph; outputting candidate root causes according to the abnormal index graph based on a random walk algorithm; and inputting the candidate root causes into a pre-trained fault diagnosis network model to verify the accuracy, thereby obtaining a service fault diagnosis result. According to the method, the service fault of the micro-service architecture can be diagnosed in real time, and the accuracy of the diagnosis result is relatively high.

Description

technical field [0001] The present invention relates to the technical field of microservice architecture service fault diagnosis, in particular to a neural network-based microservice fault diagnosis method and a storage medium. Background technique [0002] As microservice architectures become more popular, the performance of microservices is critical, as microservice failures can degrade user experience and result in financial losses. Effectively locating the root cause of a failure can help restore service and mitigate losses. In the microservice architecture, an application is often decomposed into multiple microservices, and the web application calls services running on different hosts and containers through the gateway (see Figure Service Call). It can be seen that there is indirect fault propagation in the topology called between services, and the problem cannot be directly located. Consequently, the high complexity and dynamics of microservice architectures make tro...

AI Technical Summary

Problems solved by technology

It can be seen that there is indirect fault propagation in the topology called between services, and the problem cannot be directly located

Therefore, the high complexity and dynamic nature of microservices architecture makes troubleshooting difficult

First of all, it is difficult to obtain a fixed service topology relationship, and this static troubleshooting method cannot be applied to the situation where service calls change frequently; even if we know the topology relationship between services, due to the indirect propagation of faults, we still lack Effective dynamic diagnosis mechanism to determine the root cause

Furthermore, time-series anomaly detection algorithms based on a single metric often fail to identify the root cause of failures because a single metric is not sufficient to characterize the anomalies occurring in various services

Images