Trust and Cluster Based Fair Share Scheduling in Multi Cloud Environment

Abstract

Multi-cloud can provide scalable resources to the large-scale applications, whereas single cloud resources may not be able to provide sufficient resources for an application. This paper aims to increase the job success rate and decrease the make span time in the multi -cloud environment. In our proposed work, two levels of scheduling are done for effective scheduling. First, the cloudlets are scheduled by the broker based on the trustworthiness of provider as well as objective function of the customer such as reduction in cost or makespan or both cost and makespan are in certain ratio. So, the cloudlets are allotted to the clustered resources present at the selected trustworthy providers. Equal partitioning and fair share scheduling is compared with various scenarios and then fair share scheduling is found to be able to provide better results in terms of makespan and job success rate. The clustering of both requests and resources of the proposed work and the experimental result ensures the proposed objective is achieved in terms of minimizing make span time and maximizing job success rate in multi cloud environment.