Combinatorial double auction-based resource allocation mechanism in cloud computing market

Abstract

The cloud computing environment may be considered as market for computing and storage resources. Providers rent their available resources in the form of Virtual Machines (VM) and charge the users accordingly. One of the challenges in this market is providing a mechanism for the allocation of resources and their pricing, such that the proper benefit of both users and providers are guaranteed. In this paper, a combinatorial double auction-based market is studied in which a broker performs the allocation of the providers’ VMs according to the users’ requests. The proposed allocation problem is formulated as an integer linear programming model aiming at maximizing the total profit of users and providers. It is proved that the proposed model satisfies the desirable properties including: truthfulness, fairness, economic efficiency and allocation efficiency. Furthermore, due to the high complexity of the proposed model, a heuristic resource allocation algorithm with a quasi linear time complexity is presented. The results of evaluations confirm the good agreement of the heuristic algorithm with the optimization model in terms of allocation performance. Moreover, simulation results using CloudSim indicate that, compared to the previous works in literature, the proposed algorithm increases the profit of providers and users and reduces the resource wastage.