LAUSR

Cloud computing is the delivery of on-demand, network accessed and easily configurable computing resources on a pay-as-you-go basis. These resources are dynamically assigned and can be scaled up and down in response to users’ needs. An increasing number of businesses are switching to cloud based computing systems for internal uses. Hence, there is a need to explore queuing theory to model systems bound by the constraint of a finite user population. In this work, a model for the analysis of request stage of finite population cloud computing systems has been presented. The losses in the system due to impatient users owing to balking and reneging in the system have also been taken into account in the model. A scheduling method to improve the load balancing of the request stage of the system has been introduced. A cost and reliability measurement scheme has been used to analyse the performance/cost trade-off associated with request stage of the system. The results obtained have been presented and compared with existing model working in the same domain. Experimental results show that improved cost saving (as high as $$\approx 91\%$$ ) and high throughput rate can be achieved by applying the proposed CCRM scheme.