A Fault Tolerant Strategy in Hybrid Cloud Based on QPN Performance Model

A hybrid cloud computing platform is the integration and utilization of services from both public and private cloud computing. The hybrid cloud platform will help scientists and businesses to leverage the scalability and cost effectiveness of the public cloud by paying only for IT resources consumed. However, resource failure and other faults may increase due to the heterogeneity and complexity of hybrid cloud computing. Such failures can result in frequent performance degradation and cause request rejection. In this paper, we use the Queuing Petri nets to model the hybrid cloud which can easily analyze the performance. Based on the real time performance analysis from the model, we propose a fault tolerant strategy for the hybrid cloud computing which consider both performance and request completion. The results simulated by QPME show the advantage of our strategy.