Non-intrusive Critical System Event Recognition and Prediction in Cloud

Abstract

The cloud computing platform provides an efficient and flexible way to offer services and computation facilities to users. However, reliability in the cloud is an important factor to measure the performance of a virtualized cloud computing platform. System failure, software failure, outside attacks, and mis-actions of virtual machines make the cloud computing platform unstable and unreliable. In order to avoid critical events affect reliability, resources, applications, and services can be scheduled around predicted failure and limit the impact. In the cloud computing platform, different virtual machines may generate number of system events. Events from different virtual machines can affect system stability together. Such mechanisms are especially important for cloud computing environment. In this paper, we propose a framework to recognize and predict system critical events that come from different virtual machines to increase system stability of cloud computing environment.