摘要
系统性能预测为有预见性的资源调度提供依据,是云平台管理的重要方面。从数据收集、数据处理和预测方法三个方面总结性能预测方法。提出采用移动平均线方法来预测服务性能的长期发展趋势。针对性能小范围波动时会频繁改变预测信号的问题,进一步改进此方法,引入标准差以有效过滤抖动信号。在亚马逊弹性计算云环境下验证了方法的有效性。
Performance prediction is necessary for intelligent resource allocation on the cloud platform. This paper analyzes present predicting methods from three aspects: data collection, data processing and predication methods. Then, it proposes an approach for long-term trend prediction using moving averages method. To better tolerate per- formance jitter in a small range, it further improves the conventional moving averages method with signal filtering mechanism using standard deviations. An experiment is exercised on Amazon Elastic Compute Cloud platform to il- lustrate the proposed approach of performance monitoring, analysis and prediction.
出处
《计算机科学与探索》
CSCD
2012年第6期495-503,共9页
Journal of Frontiers of Computer Science and Technology
基金
国家自然科学基金No.61073003
国家重点基础研究发展规划(973)No.2011CB302505~~
关键词
性能预测
云计算
移动平均线
performance prediction
cloud computing
moving averages