一种不需协商的业务等级协商测量方法

Method for service level agreement measurement without negotiation

提出了一种针对业务等级协商(SLA)数据特性的模式提取方法。本方法主要抽取出系统的关键性能指标和关键质量指标,根据用户的体验进行机器学习发现模式,利用网格理论有效地避免了求多个指标之间关联度的问题,制定出一种实际运行中可以自动归纳总结满足业务SLA策略的数据度量方法,并且利用可视化数据展示的方式让用户很容易辨别得到的模式是否可信,从而调整系统的输入参数以便得到更加准确的结果。

A novel method was proposed to extract SLA patterns from large data sets. This method extracts patterns from large data sets of customers＇ experience through machine learning according to key performance indicators and key quality indicators predefined. By learning, Telecommunication Company can control the allocation of resource for holding customers＇ requirements from low level to high level. It need not consider the associations of multi-variants through grid partition. What＇s more, this method can discern the accuracy of extracted patterns through visualization technologies in order to adjust user-specific input parameters.