QoS and cost aware brokerage system for selecting green CDCs

Abstract

Cloud service brokerage (CSB) systems are getting increasingly popular day by day to manage the performance and utility between cloud service providers (CSP) and cloud service consumers (CSC). An optimized CSB is supposed to be cost effective, energy efficient and should maintain the quality of service (QoS) according to the service level agreement (SLA). Additionally, a CSB needs to have the capability to select the optimal CSP which reflects the requirements of user requests. The state-of-the-art works on establishing a CSB system did not consider all the three objectives that a smart CSB system demands. In this work, we have defined a CSB model based on all of these three necessary criteria to achieve an optimized CSB which are energy efficiency, cost effectiveness and certainty of achieving QoS. The Mixed-Integer Linear Program (MILP) is used to formulate the objective function that will select a CSB. We simulate our proposed model in Matlab, compare our model with existing working models and the results show that significant performance improvement is achieved by our newly defined CSB.