论文部分内容阅读
随着越来越多的在线业务被迁移到基于云的平台上,如何检测云平台上在线业务的异常运行状态成为了一个重要的问题。现有方法通过分析在线业务的实时负载数据来判断业务是否存在异常,在应对由程序异常或突发用户访问引起的异常负载时存在准确率低、误报率高的问题。该文提出并实现了一种面向云计算在线业务的异常负载检测方法。该方法利用小波分析技术,将原始负载数据分解成频率不同的多条曲线,并利用统计分析技术,通过检测各个频率上的异常增长或降低来判断负载是否存在异常。实验结果表明:同现有方法相比,该方法更准确,同时可以大大降低误报率。
As more and more online businesses are migrated to cloud-based platforms, how to detect the abnormal operation of online businesses on cloud platforms has become an important issue. In the existing method, the real-time load data of the online service is analyzed to judge whether the service is abnormal or not. When the abnormal load caused by the program exception or the unexpected user access exists, the existing method has the problems of low accuracy and high false alarm rate. This paper proposes and implements an abnormal load detection method for online business of cloud computing. The method uses wavelet analysis technology to decompose the original load data into multiple curves with different frequencies and use statistical analysis techniques to determine whether there is an abnormal load by detecting the abnormal increase or decrease of each frequency. The experimental results show that this method is more accurate than the existing methods and can greatly reduce the false alarm rate.