A queuing model considering resources sharing for cloud service performance

摘要

Cloud computing is a novel paradigm for the provision of service on demand. Performance analysis of cloud computing is required to predict the corresponding quality of service experienced by users. Although there are some researches on cloud service performance, very few of them considered the impact of resources sharing among virtual machines (VMs). This paper presents a queuing model for performance analysis of cloud services. The model considers the resources sharing among VMs. The service requests are relaxed compared with prior research. A service request is divided into many subtasks and each subtask is served by a VM. Multiple VMs share the same underlying physical resources. The performance indicators such as the average response time, blocking probability are obtained, and a numerical example is presented.