Dynamic and Static Characteristics Based Algorithm to Allocate VMs to Jobs in the Cloud
Ellendula Madhukar, Thirumalaisamy Ragunathan · 2016
In present scenario, many organizations are trying to minimize the capital expenditure by using the cloud. In the cloud, IT infrastructure and applications are provided as services based on pay-as-you-use model. Cloud providers try to achieve maximum profits in short time, while the cloud users want their work to be done with minimum cost. However, utilization of cloud resources efficiently is an NP-hard optimization problem. Many algorithms have been proposed in the literature for scheduling VMs in cloud environment. The disadvantage with these algorithms is the high time complexity. In this we have proposed a new algorithm which analyses logs, where details of the jobs are stored, for predicting the execution time of the new job. The proposed algorithm schedules the job to a particular VM, based on the average turn around time and other dynamic properties of the load existing in the VMs, present in the cloud. Our preliminary study indicates that our algorithm is able to provide good performance by reducing the time complexity.