面向容器化PaaS平台的智能监控技术研究与实现

发布时间：2018-05-01 21:01

本文选题：云计算 + 平台即服务　；参考：《计算机集成制造系统》2017年05期

【摘要】：为了弥补现有的容器化PaaS平台监控方式在监控粒度和深度上的不足,研究并实现了一种智能监控方法。在该方法中,监控模块通过动态调整采样周期,使采样点集中于应用系统资源占用状况波动幅度较大的时段;对于波动幅度超过阈值的情况,采集应用系统调用级别的数据并加以分析;将统计分析结果及原始监测数据按时序整合并对外提供查询接口,进而为用户定位异常或发现性能瓶颈提供帮助。相较于现有的监控方案,该方法能更加细粒度地反映应用运行时状态的变化,同时动态的数据采集机制能有效节省计算和存储资源的开销。
[Abstract]:In order to make up for the deficiency of the existing monitoring methods of container PaaS platform in monitoring granularity and depth, an intelligent monitoring method is studied and implemented. In this method, the monitoring module dynamically adjusts the sampling period to make the sampling point concentrate on the period in which the resource occupation of the application system fluctuates greatly, and for the situation where the fluctuation amplitude exceeds the threshold, The data of application system call level are collected and analyzed, and the statistical analysis results and the original monitoring data are integrated according to the time sequence and the query interface is provided to help the user locate the exception or find the performance bottleneck. Compared with the existing monitoring scheme, this method can reflect the change of the application runtime state more fine-grained, and the dynamic data acquisition mechanism can effectively save the cost of computing and storage resources.
【作者单位】：浙江大学计算机科学与技术学院;衢州职业技术学院信息工程学院;
【基金】：浙江省重大科技专项资助项目(2015C01027) 浙江省自然科学基金资助项目(LY17F020014)~~
【分类号】：TP277;TP393.09

【相似文献】