A Novel Cloud Monitoring Framework with Enhanced QoS Supporting

Abstract

In Cloud environments, performance monitoring service is to obtain a full knowledge of underlying resources. However, it is still a challenging task to manage the information of heterogeneous resources in an efficient way, which is especially true when user-specific quality-of-service (QoS) should be concerned. Although many Cloud monitoring solutions have been proposed in recent years, most of them only passively raise an alert event when a QoS violation occurs. In this article, the authors present a novel Cloud monitoring framework, in which enhanced-QoS is supported through three mechanisms: proactive service-layer agreement (SLA) violation prediction, SLA ranking service, and multi-tenant resource monitoring mechanism. Extensive experiments are conducted in a realistic cloud platform, and the results indicate the proposed framework is capable of providing better QoS supporting comparing with existing monitoring solutions. Also, it exhibits desirable scalability and adaptiveness in a wide range of experimental scenarios.