微型电脑应用2025，Vol.41Issue(4)：202-206,232,6.
基于Prometheus的云平台多容器集群异常监控技术
Cloud Platform Multi-container Cluster Anomaly Monitoring Technology Based on Prometheus
戚林成 1陶笑 1冷新云 1韦健1
作者信息
- 1. 江苏电力信息技术有限公司,江苏,南京 210024
- 折叠
摘要
Abstract
The cloud platform's multi-container cluster has a large amount of data and involves multiple types,making it difficult to monitor abnormal states.Therefore,a monitoring algorithm based on Prometheus is proposed.In the cloud platform,the wavelet decomposition method is used to obtain the real-time state sequence of multi-container cluster data,and binary tree de-composition description method is used to divide the characteristics of different types of cluster data.The monitoring space is divided based on the distributed storage management characteristics of Prometheus technology,and a monitoring class center can be obtained.By comparing the similarity between multi container cluster data and the node center,the data with the stron-gest similarity are abnormal.The simulation experiment shows that the proposed method monitors the abnormal state data in-trusion signal,which changes significantly between 800～1200 test points,and is very close to the actual number format excep-tion(NFE)abnormal state data intrusion monitoring results.The proposed method has a low CPU consumption rate,the mini-mum value is 15％,and the average response time to the abnormal monitoring is 1.7 s,which can help the cloud platform to operate stably.关键词
多容器集群/数据监控/二叉树分解描述法/小波分解法/状态序列Key words
multi-container cluster/data monitoring/binary tree decomposition description method/wavelet decomposition method/status sequence分类
信息技术与安全科学引用本文复制引用
戚林成,陶笑,冷新云,韦健..基于Prometheus的云平台多容器集群异常监控技术[J].微型电脑应用,2025,41(4):202-206,232,6.
