Percentile performance analysis of infrastructure-as-a-service clouds with task retrials

Abstract

Performance evaluation of cloud infrastructures and cloud-based applications is required to evaluate and quantify the cost-benefit of a strategy portfolio and the quality of service (QoS) experienced by end-users. For this purpose, we introduce an analytical framework to percentile-based performance analysis of unreliable Infrastructure-as-a-Service clouds with faulty and retrial tasks. The performance measured in a certain level percentile of the instantiation time predicted given variable load intensities, fault frequencies, multiplexing abilities, and instantiation processing delays. A case study based on a real-world campus cloud is carried out to show the correctness of the proposed theoretical model.