On the design of reactive approach with flexible checkpoint interval to tolerate faults in cloud computing systems

摘要: The likelihood of failures rises in cloud computing systems as a result their unstable nature. Additionally, the size system varies with time and thus become common incident. Failures have high impact on performance expected benefits for both customers providers. Fault tolerance is an essential challenge facing providers order to mitigate effects maintaining Service Level Agreement (SLA) satisfied. Checkpointing one most known reactive fault techniques used distributed computing. However, it can incur considerable overheads that depend interval checkpoint applied these put down cloud. In this paper, approach context checkpointing proposed evaluated aim getting better performance. depends applying flexible reduce overheads. Simulation experiments indicate superior terms power consumption, response time, monetary cost capacity.